Multi-time-scale optimization method and system for integrated energy systems

Abstract

The present invention relates to the field of integrated energy dispatch optimization, specifically to a multi-time-scale optimization method, system, equipment, and medium for an integrated energy system, including: constructing a system multi-objective robust optimization configuration model based on renewable energy output and user load forecast results; solving the system multi-objective robust optimization configuration model using a multi-objective genetic algorithm to determine the optimal system robust structural configuration; dividing the control process of the integrated energy system into three time stages, and optimizing the three time stages according to the optimal system robust structural configuration. Considering the most unfavorable source-load situation, the integrated energy system is robustly optimized and configured. The resulting equipment capacity result is relatively large, but it can meet the load of the park as much as possible, and tries not to rely on the external large power grid or large heating network, reducing the interaction between the outside world and the park's integrated energy system, improving the independence of the system's park-level energy system, and improving the park's near-zero energy consumption level.

Description

Technical Field

[0001] The present invention relates to the field of integrated energy scheduling optimization, and specifically to a multi-time scale optimization method and system for an integrated energy system. Background Art

[0002] Most current park energy supply systems are designed solely to meet the park's cooling, heating, and electricity loads, without considering the park's independence or the goal of achieving near-zero energy consumption. The current trend in integrated energy systems is to build energy supply systems that meet the park's loads while maintaining a degree of independence from the main power grid and heating network, thereby reducing external energy input.

[0003] The park-level integrated energy system fully utilizes renewable energy for energy supply. However, renewable energy is intermittent, decentralized, and volatile, and the load of park users is affected by factors such as the environment, personnel behavior, and pricing policies, resulting in uncertainty in the energy supply and demand on both the source and load sides. In addition, the operating characteristics of various devices in the integrated energy system vary greatly, including "fast-response" equipment related to electricity and "slow-response" equipment related to cooling and heating. The operating characteristics of "fast-response" and "slow-response" devices need to be fully coordinated to meet users' real-time energy needs and instantaneous load changes.

[0004] Defects and deficiencies of existing technologies: (1) In the process of optimizing the integrated energy system, existing technologies only consider multiple target requirements and the balance between source and load supply and demand, and optimize the large power grid as a backup energy source instead of performing robust optimization. This results in a large amount of grid interaction between the power grid and the park system, and reduces the independence of the park system. However, the large power grid does not want to absorb the surplus power of the park, which will affect its own power quality. The park system also does not want to purchase too much power from the large power grid, resulting in excessive dependence on the large power grid. (2) The control mode of existing integrated energy systems is generally based on electricity to determine heat or heat to determine electricity, or based on the operation mode obtained by multi-objective optimization. It does not consider the source and load energy mismatch caused by multiple uncertain factors such as randomness, volatility, and intermittency of source and load energy. That is, the existing literature only considers the system's day-ahead optimal scheduling, and does not consider the source and load conditions at the actual operation time of the system, that is, it does not consider the operation mode of the system within the day. (3) Existing technologies give little consideration to multi-energy storage devices and power-to-heat and power-to-cooling methods. Even if some projects consider multiple energy storage methods such as electricity storage, heat storage, and cold storage, there is no collaborative operation model between different energy storage methods. Instead, the surplus electricity is simply stored in the electricity storage device, and the electricity is released when the user's electricity load cannot be met; the surplus thermal energy is stored in the heat storage device, and the heat is released when the user's heat load cannot be met; the surplus cold energy is stored in the cold storage device, and the cold is released when the user's cold load cannot be met. There is no collaborative operation optimization between them. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method, system, device and medium for robust optimization configuration of an integrated energy system to solve the problem that functional systems cannot achieve zero energy consumption.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a multi-time-scale optimization method for an integrated energy system, comprising:

[0008] Based on the forecast results of renewable energy output and user load, a multi-objective robust optimization configuration model for the system is constructed;

[0009] The multi-objective genetic algorithm is used to solve the system multi-objective robust optimization configuration model and determine the optimal system robust structure configuration;

[0010] The regulation process of the integrated energy system is divided into three time stages, which are then optimized according to the optimal system robust structure configuration.

[0011] Furthermore, a multi-objective robust optimization configuration model of the system is constructed, including objective functions and constraints;

[0012] The objective functions include: minimum carbon emissions, minimum annualized total cost, maximum primary energy utilization rate, and maximum independence of the integrated energy system;

[0013] The constraints include: energy supply and demand balance, and equipment capacity being less than or equal to the equipment maximum capacity constraint.

[0014] Furthermore, the energy supply and demand balance includes: electricity balance constraint, heat balance constraint and cooling balance constraint;

[0015] Electricity balance constraint: E ICE (t)+E grid (t)+P pv (t)+P Battery，out (t)+P WT (t)≥E build (t)+E ASHP (t)+E HP (t); where t is hourly data; E ICE is the power generation of the gas turbine, kWh; E grid is the amount of electricity purchased from the municipal power grid, kWh; P pv is photovoltaic power generation, kwh; P Battery，out is the battery discharge capacity, kWh; P WT is the discharge capacity of wind power generation, kWh; E build is the park electricity load, kWh; E ASHP is the power consumption of the electric refrigerator, kWh; EHP is the power consumption of the ground source heat pump, kWh;

[0016] Heat balance constraint: Q ex (t)+Q GB (t)+Q solar (t)+P HTS，out (t)≥Q build (t); where Q ex is the heat release of the heat exchanger, kWh; Q GB Heat released by the boiler, kWh; Q solar is the heat collected by the solar thermal system, kWh; P HTS，out Heat released to the heat storage system, kWh; Q build is the park heat load, kWh;

[0017] Cooling balance constraint: C ASHP (t)+C HP (t)+C AHP (t)≥C build (t), where C ASHP is the cooling capacity of the electric refrigerator, kWh; C HP is the cooling capacity of the ground source heat pump, kWh; C AHP is the cooling capacity of the absorption refrigerant, kWh; C build is the cooling load of the park, kWh.

[0018] Furthermore, the device capacity is less than or equal to the device maximum capacity constraint includes: G Battery，in ≤G Battery，in，max , G Battery，out ≤G Battery，out，max , G HST，in ≤G HST，in，max , G HTS，out ≤G HTS，out，max , G pv ≤G pv，max , G solar ≤G solar，max , G ICE ≤G ICE，max , G ASHP ≤G ASHP，max , G GB ≤G GB，max , G WT ≤G WT，max , G HP ≤G HP，max , G AHP ≤G AHP，max , where G represents the capacity of the equipment, kW; G Battery，tn is the charging capacity of the battery, kW; G Battery，out is the discharge capacity of the battery, kW; G HST，inis the heat storage capacity of the heat storage system, kW; G HTS，out is the heat release capacity of the heat storage system, kW; G pv is the capacity of the photovoltaic system, kW; G solar is the capacity of the solar thermal system, kW; G ICE is the capacity of the internal combustion engine, kW; G ASHP is the capacity of the electric refrigerator, kW; G GB is the capacity of the gas boiler, kW; G WT is the capacity of the wind turbine, kW; G HP is the capacity of the ground source heat pump, kW; G AHP is the capacity of the absorption refrigerant, kW.

[0019] Furthermore, the minimum carbon emission is minC yx =min(C yx，hear +C yx，cold +C yx，ele ), where C yx is the total carbon production during the entire life cycle of the park system operation, tCO2; C yx，hear The carbon production of the park system during the whole life cycle of providing heat to users, tCO2; C yx，cold The carbon production of the entire life cycle of the park system in the process of providing cooling to users, tCO2; C yx，ele The carbon production of the park system during its entire life cycle in the process of supplying power to users, tCO2.

[0020] Furthermore, the minimum annualized total cost is min(C cost )=min(∑ i A i ×c inv，i ×G eq，i ×n i )+∑ i (A i ×c ope，i ×n i ), Where i is the device category; C cost is the annual total cost of the system, yuan; A i is the cost recovery coefficient; c inv is the initial investment cost per unit capacity of the equipment, RMB / kW; G eq is the equipment capacity, kW; n is the number of equipment; r is the interest rate; y i Indicates the life cycle of the equipment, years; c ope It is the equipment operation and maintenance cost coefficient, including system maintenance cost and fuel consumption cost.

[0021] Furthermore, the primary energy utilization rate is the highest: Among them, Puser is the user's annual cooling, heating and electricity consumption, kWh; P con is the total energy consumed by the system, kWh.

[0022] Furthermore, the highest independence of the integrated energy system is: Among them, P grid，buy It represents the total amount of electric energy purchased and sold by the system from the external large power grid and large thermal network, kWh; P grid，sell It represents the total amount of heat energy purchased and sold by the system from the external large power grid and large thermal network, kWh; P load Refers to the total amount of cooling, heating and electricity load of the user, kWh.

[0023] Furthermore, the three time stages include: long time scale stage, medium time scale stage and short time scale stage;

[0024] The constraints in the long time scale stage and the medium time scale stage include: power balance constraint, heat balance constraint, cooling balance constraint, and equipment output being less than or equal to the maximum capacity constraint of the equipment;

[0025] The constraints in the short time scale phase include: power balance constraint and equipment output being less than or equal to the equipment maximum capacity constraint;

[0026] The output of the equipment is less than or equal to the equipment capacity constraint: P Battery，tn ≤G Battery，in , P Battery，out ≤G Battery，out , P HST，in ≤G HST，in , P HTS，out ≤G HTS，out , P ICE ≤G ICE , P ASHP ≤G ASHP , P GB ≤G GB , P HP ≤G HP , P AHP ≤G AHP , where P represents the output of the equipment, kW; G represents the capacity of the equipment, kW; P Battery，in Indicates the charging output of the battery; G Battery，in is the charging capacity of the battery, kW; P Battery，out The discharge output of the battery; G Battery，out is the discharge capacity of the battery, kW; P HST，in is the heat storage output of the heat storage system, kW; G HST，in is the heat storage capacity of the heat storage system, kW; P HTS，out is the heat release output of the heat storage system, kW; G HTS，outis the heat release capacity of the heat storage system, kW; P ICE is the output of the internal combustion engine, kW; G ICE is the capacity of the internal combustion engine, kW; P ASHP is the output of the electric refrigerator, kW; G ASHP is the capacity of the electric refrigerator, Kw; P GB is the output of the gas boiler, Kw; G GB is the capacity of the gas boiler, kW; P HP is the output of the geothermal heat pump, kW; G HP is the capacity of the ground source heat pump, kW;

[0027] The output of the electric energy equipment is less than or equal to the equipment capacity constraint: P Battery,in ≤G Battery,in ,P Battery,out ≤G Battery,out, , P ASHP ≤G ASHP ,P HP ≤G HP , where P represents the output of the equipment, kW; G represents the capacity of the equipment, kW; P Battery,in Indicates the charging output of the battery; G Battery,in is the charging capacity of the battery, kW; P Battery,out The discharge output of the battery; G Battery,out is the discharge capacity of the battery, kW; P ASHP is the output of the electric refrigerator, kW; G ASHP is the capacity of the electric refrigerator, Kw; P HP is the output of the geothermal heat pump, kW; G HP is the capacity of the ground source heat pump, kW;

[0028] The scheduling optimization objective in the long time scale stage is Among them, C D h is the system operation cost, yuan; T is the scheduling period, 24h; is the interaction cost between CCHP microgrid system and grid, RMB; is the natural gas fee, RMB; is the battery aging cost, yuan; is the system operation and maintenance cost, RMB; is the environmental cost of the system, yuan;

[0029] The scheduling optimization objective of the medium time scale stage is Where, M is the time span, 4h; is the cost of grid interaction regulation, yuan; is the fuel adjustment cost, yuan; Penalty cost for battery charge and discharge power variation, RMB; Penalty cost for heat storage / cold tank heat storage and release power change, RMB;

[0030] The short-time-scale stage scheduling optimization objective is in, is the real-time power purchase adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time boiler power adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time gas purchase power adjustment, kW; is the real-time power adjustment of the absorption chiller, kW; is the real-time power adjustment of the electric refrigerator, kW; is the real-time power adjustment of the ground source heat pump, kW.

[0031] In a second aspect, the present invention provides a multi-time-scale optimization system for an integrated energy system, comprising:

[0032] Model building module: used to build a system multi-objective robust optimization configuration model based on renewable energy output and user load forecast results;

[0033] Computational solution module: used to solve the system's multi-objective robust optimization configuration model using a multi-objective genetic algorithm to determine the optimal system robust structural configuration;

[0034] Configuration optimization module: used to divide the integrated energy system into three time stages and optimize the three time stages according to the optimal system robust structure configuration.

[0035] The present invention has at least the following beneficial effects:

[0036] 1. The present invention provides a multi-time-scale optimization method for an integrated energy system, including: constructing a system multi-objective robust optimization configuration model based on renewable energy output and user load forecast results; solving the system multi-objective robust optimization configuration model using a multi-objective genetic algorithm to determine the optimal system robust structural configuration; dividing the control process of the integrated energy system into three time stages, and optimizing the three time stages according to the optimal system robust structural configuration. Considering the most unfavorable source-load situation, the integrated energy system is robustly optimized and configured. The resulting equipment capacity result is relatively large, but it can meet the park load as much as possible, and try not to rely on external large power grids and large heating networks, reduce the interaction between the outside world and the park integrated energy system, improve the independence of the system park-level energy system, and improve the park's near-zero energy consumption level;

[0037] 2. The present invention provides a multi-time scale optimization method for an integrated energy system, wherein the three time stages include: a long time scale stage, a medium time scale stage, and a short time scale stage; the constraints of the long time scale stage and the medium time scale stage include: power balance constraint, heat balance constraint, cooling balance constraint, and equipment output being less than or equal to the maximum capacity constraint of the equipment; the constraints of the short time scale stage include: power balance constraint and power equipment output being less than or equal to the maximum capacity constraint of the equipment; the scheduling optimization objective of the long time scale stage is Among them, C D h is the system operation cost, yuan; T is the scheduling period, 24h; is the interaction cost between CCHP microgrid system and grid, RMB; is the natural gas fee, RMB; is the battery aging cost, yuan; is the system operation and maintenance cost, RMB; is the environmental cost of the system, yuan; the scheduling optimization objective of the medium time scale stage is Where, M is the time span, 4h; is the cost of grid interaction regulation, yuan; is the fuel adjustment cost, yuan; Penalty cost for battery charge and discharge power variation, RMB; is the penalty cost of heat storage / cold tank heat storage and release power change, RMB; the short time scale stage scheduling optimization objective is

[0038] in, is the real-time power purchase adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time boiler power adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time gas purchase power adjustment, kW; is the real-time power adjustment of the absorption chiller, kW; is the real-time power adjustment of the electric refrigerator, kW; is the real-time power adjustment of the ground-source heat pump, kW. Using a time interval approach, the authors non-manually set the scheduling interval. This approach comprehensively considers the differences in the operating characteristics of each device and the system's energy supply quality to determine the optimal scheduling interval for each stage. A "long time scale, medium time scale, short time scale" optimization scheduling scheme is proposed. This scheme also considers the operating characteristics of "fast-response" devices related to electricity and "slow-response" devices related to cooling and heating, as well as the randomness, volatility, and intermittency of both the source and load sides. A multi-time scale optimization operation method for a park-level integrated energy system is proposed. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings, which constitute part of the present invention, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0040] Figure 1 Schematic diagram of the multi-time-scale optimization operation method of the energy supply system of the present invention;

[0041] Figure 2 Optimize the control flow chart for the system of the present invention;

[0042] Figure 3 This is a structural block diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0043] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other.

[0044] The following detailed description is an exemplary description and is intended to provide further detailed description of the present invention. Unless otherwise indicated, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present invention belongs. The terms used in the present invention are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention.

[0045] Example 1

[0046] The present invention provides a method for robust optimization configuration of an integrated energy system, comprising:

[0047] Based on the forecast results of renewable energy output and user load, a multi-objective robust optimization configuration model for the system is constructed;

[0048] The multi-objective genetic algorithm is used to solve the system multi-objective robust optimization configuration model and determine the optimal system robust structure configuration;

[0049] The regulation process of the integrated energy system is divided into three time stages, which are then optimized according to the optimal system robust structure configuration.

[0050] Construct a typical architecture for a park-level integrated energy system and conduct multi-objective robust optimization of the system. The specific steps are as follows:

[0051] (1) Based on the forecast results of renewable energy output and user load, convert them into deterministic results.

[0052] Fluctuations in environmental factors like temperature, humidity, and wind speed make it difficult to accurately predict the amount of renewable energy. Furthermore, human and market factors can cause load demand to fluctuate. Combining robust optimization, the source-load energy can be described as a robust uncertainty set.

[0053] The uncertainty is described as follows: Among them, P RE is the actual value of renewable energy output, is the nominal value of renewable energy output, L is the actual load value, and L 0 is the nominal load value, ΔP RE is the maximum fluctuation of renewable energy output, and ΔL is the maximum fluctuation of load. Considering that the output of renewable energy and load demand do not always reach the worst case in every period, the uncertainty factor can be further described as: Among them, T RE , T represent the renewable energy output or user load time period, Γ RE Γ and Γ are respectively robust measures of renewable energy output and user load, which can be integers or non-integers. In a finite time period, the robust measure Γ represents the maximum amount that the uncertainty factor deviates from the nominal value, that is, renewable energy output or load demand cannot experience the worst-case random fluctuations simultaneously in all time periods. If Γ is 0, there is no random fluctuation in renewable energy output and load demand, and the corresponding true value is equal to the nominal value, that is, a deterministic situation; if Γ RE If 7 is selected, it means that the output of renewable energy is between 0 and T RE The worst randomness occurs in any 7 periods of time. Similarly, the worst randomness occurs in any 7 periods of time from 0 to T for the cooling, heating and electricity load demands. If Γ RE and Γ take T RE and T, indicating that the renewable energy output and load demand exhibit the worst-case stochastic conditions in each time period, resulting in the optimal solution having the highest degree of conservatism. Controlling the value of the robust measure not only allows for a realistic description of the stochastic scenario but also regulates the cost and conservatism of robust optimization. Therefore, robust measures of renewable energy output and load demand are important regulators of the conservatism level of the economic operation of integrated energy systems.

[0054] (2) Constructing a multi-objective robust optimization configuration model for the system

[0055] Considering multiple objectives such as economic efficiency, environmental protection, energy efficiency, and independence, a multi-objective robust optimization configuration model is constructed for the system, targeting the worst-case scenario of renewable energy output and user cooling, heating, and electricity loads over the system's entire life cycle. The optimization variable is the capacity of each device in the campus-level integrated energy system. The constructed integrated energy system must meet the following constraints:

[0056] ①Energy supply and demand balance equation

[0057] 1) Electricity balance constraint: E ICE (t)+E grid (t)+P pv (t)+P Battery，out (t)+P WT (t)≥E build (t)+E ASHP (t)+E HP (t); where t is hourly data; E ICE is the power generation of the gas turbine, kWh; E grid is the amount of electricity purchased from the municipal power grid, kWh; P pv is photovoltaic power generation, kwh; P Battery，out is the battery discharge capacity, kWh; P WT is the discharge capacity of wind power generation, kWh; E build is the park's electricity load, kWh; E ASHP is the power consumption of the electric refrigerator, kWh; E HP is the power consumption of the ground source heat pump, kWh;

[0058] 2) Heat balance constraint: Q ex (t)+Q GB (t)+Q solar (t)+P HTs，out (t)≥Q build (t); where Q ex is the heat release of the heat exchanger, kWh; Q GB Heat released by the boiler, kWh; Q solar is the heat collected by the solar thermal system, kWh; P HTS，out Heat released to the heat storage system, kWh; Q build is the park heat load, kWh;

[0059] 3) Cooling balance constraint: C ASHP (t)+C HP (t)+C AHP (t)≥C build (t), where C ASHP is the cooling capacity of the electric refrigerator, kWh; C HP is the cooling capacity of the ground source heat pump, kWh; C AHP is the cooling capacity of the absorption refrigerant, kWh; C build is the cooling load of the park, kWh.

[0060] ② Equipment capacity ≤ equipment maximum capacity constraint: G Battery，in ≤G Battery，in，max , G Battery，out ≤G Battery，out，max , GHST，in ≤G HST，in，max , G HTS，out ≤G HTS，out，max , G pv ≤G pv，max , G solar ≤G solar，max , G ICE ≤G ICE，max , G ASHP ≤G ASHP，max , G GB ≤G GB，max , G WT ≤G wT，max , G HP ≤G HP，max , G AHP ≤G AHP，max , where G represents the capacity of the equipment, kW; G Battery，in is the charging capacity of the battery, kW; G Battery，out is the discharge capacity of the battery, kW; G HST，in is the heat storage capacity of the heat storage system, kW; G HTS，out is the heat release capacity of the heat storage system, kW; G pv is the capacity of the photovoltaic system, kW; G solar is the capacity of the solar thermal system, kW; G ICE is the capacity of the internal combustion engine, kW; G ASHP is the capacity of the electric refrigerator, kW; G GB is the capacity of the gas boiler, kW; G WT is the capacity of the wind turbine, kW; G HP is the capacity of the ground source heat pump, kW; G AHP is the capacity of the absorption refrigerant, kW.

[0061] (3) Solve the system model using a multi-objective genetic algorithm to determine the optimal system robust structure configuration. The objective functions include: minimizing carbon emissions generated during the energy consumption of the park throughout its life cycle, and annualized total cost ( cost , yuan) is minimized, the system's primary energy utilization rate (R) is maximized, and the park-level integrated energy system has the highest independence. The calculation methods for each objective function are shown below, and the weights of each objective are determined using the TOPSIS method.

[0062] ① The carbon emissions generated by the park's energy consumption during the entire life cycle are minimal (including carbon emissions generated by electricity purchases and natural gas consumption)

[0063] mainC yx =min(Cyx，hear +C yx，cold +C yx，ele ), where C yx is the total carbon production during the entire life cycle of the park system operation, tCO2; C yx，hear The carbon production of the park system during the whole life cycle of providing heat to users, tCO2; C yx，cold The carbon production of the entire life cycle of the park system in the process of providing cooling to users, tCO2; C yx，ele The carbon production of the park system during its entire life cycle in the process of supplying power to users, tCO2.

[0064] ②Annualized total cost (C cost , yuan) minimum (including system initial investment cost and operation and maintenance cost)

[0065] min(C cost )=min(∑ i A i ×c inv，i ×G eq，i ×n i )+∑ i (A i ×c ope，i ×n i ), Where i is the device category; C cost is the annual total cost of the system, yuan; A i is the cost recovery coefficient; c inv is the initial investment cost per unit capacity of the equipment, RMB / kW; G eq is the equipment capacity, kW; n is the number of equipment; r is the interest rate; y i Indicates the life cycle of the equipment, years; c ope It is the equipment operation and maintenance cost coefficient, including system maintenance cost and fuel consumption cost.

[0066] ③The primary energy utilization rate (R) of the system is the highest

[0067] Among them, P user is the user's annual cooling, heating and electricity consumption, kWh; P con is the total energy consumed by the system, kWh.

[0068] ④The park-level integrated energy system has the highest independence

[0069] Among them, P grid，buy It represents the total amount of electric energy purchased and sold by the system from the external large power grid and large thermal network, kWh; P grid，sell It represents the total amount of heat energy purchased and sold by the system from the external large power grid and large thermal network, kWh; P loadRefers to the total amount of cooling, heating and electricity load of the user, kWh.

[0070] The scheduling time interval between “medium time scale” and “short time scale” (Δt m ) Determination method: ①Δt m ≥Δt 2 , take each device Δt m The minimum value is taken as the optimal Δt of the system m , reduce the number of equipment dispatch; ② Consider the worst-case scenario for environmental conditions. To ensure the quality of energy supply, Δt m Not too large; ③ Considering the thermal inertia of the building itself, Δt m Appropriate increase; ④ Due to the load fluctuation frequency of users in different time periods, different scheduling time intervals are set. Therefore, the time interval formula is:

[0071] Where Δt zl Indicates the upper limit of the scheduling time interval (taking into account the energy supply quality); Δt dx represents the increase in the scheduling time interval (taking into account thermal inertia); j represents different equipment; f load Indicates load fluctuation frequency; f ref Indicates the reference value of the fluctuation frequency.

[0072] The control process of the integrated energy system is divided into three stages: "long time scale", "medium time scale", and "short time scale". Through step-by-step coordination, the impact of uncertainty on the system operation is gradually reduced. Among them, all equipment participates in the optimization operation of the "long time scale" and "medium time scale" stages, while only electricity-related fast-response equipment participates in the "short time scale". Regarding the time scale stage, the present invention provides an overall concept divided into: long time scale, medium time scale and short time scale;

[0073] (1) “Long time scale” stage

[0074] The constraints in the long-term scale stage include: electricity balance constraint, heat balance constraint, cooling balance constraint and equipment output less than or equal to the maximum capacity constraint of the equipment. It is executed once a day (resolution is 1h). Based on the forecast data of renewable energy and user cooling, heating and electricity loads on the day before, the "long-term scale" operation plan is determined to obtain the start and stop status, output plan, etc. of each device for the next day. The optimization variables in this stage are the hourly output power and start and stop status of each type of equipment. The objective function is to minimize the daily operating cost. The day-ahead operating cost mainly includes grid interaction costs, natural gas purchase costs, battery aging costs, equipment operating costs and environmental costs. The day-ahead scheduling optimization goal is: Among them, C D j is the system operation cost, yuan; T is the scheduling period, 24h; is the interaction cost between CCHP microgrid system and grid, RMB; is the natural gas fee, RMB; is the battery aging cost, yuan; is the operation and maintenance cost of the system, RMB; is the environmental cost of the system, yuan.

[0075] (2) “Medium time scale” stage

[0076] The constraints in the medium time scale stage include: power balance constraint, heat balance constraint, cooling balance constraint and equipment output less than or equal to the maximum capacity constraint. m , 1 is executed once (time span is 4 hours). Based on short-term source-load forecast data, the equipment's output plan for the "long time scale" phase is adjusted, and the "medium time scale" operation plan is determined. The optimization variable in this phase is the change in equipment power adjustment within the day, and the objective function is to minimize the power purchase cost and the penalty cost of energy storage output change within the rolling time domain. Where M is the time span, 4h; is the regulation cost of grid interaction, RMB; is the fuel adjustment cost, yuan; Penalty cost for battery charge and discharge power variation, RMB; Penalty cost for heat storage / cold tank heat release power change, RMB.

[0077] (3) “Short time scale” stage

[0078] The constraints in the short time scale phase include: power balance constraint and power equipment output less than or equal to the maximum capacity constraint. m，2 Execute once. Based on ultra-short-term source and load forecast data and the energy supply system's "medium-timescale" phase plan, determine the output of "fast-response" equipment. The optimization variable in this phase is the real-time power adjustment of the equipment. The objective function is to minimize the total adjustment of controllable equipment within the next period (e.g., 15 minutes) at the beginning of each real-time period (e.g., every 5 minutes). in, is the real-time power purchase adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time boiler power adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time gas purchase power adjustment, kW; is the real-time power adjustment of the absorption chiller, kW; is the real-time power adjustment of the electric refrigerator, kW; is the real-time power adjustment of the ground source heat pump, kW.

[0079] The “long time scale” scheduling plan is made once every day at 24:00, and at the same time every Δt m，1 Roll out a "medium time scale" scheduling plan, every Δt m，2 A "short-time scale" scheduling plan is formulated on a rolling basis. As time goes by, the time periods corresponding to the "medium-time scale" and "short-time scale" scheduling plans continue to move forward.

[0080] Equipment output is less than or equal to equipment capacity constraint: P Battery，in ≤G Battery，in , P Battery，out ≤G Battery，out , P HST，in ≤G HST，in , P HTS，out ≤G HTS，out , P ICE ≤G ICE , P ASHP ≤G ASHP , P GB ≤G GB , P HP ≤G HP , P AHP ≤G AHP , where P represents the output of the equipment, kW; G represents the capacity of the equipment, kW; P Battery，in Indicates the charging output of the battery; G Battery，in is the charging capacity of the battery, kW; P Battery，out The discharge output of the battery; G Battery，out is the discharge capacity of the battery, kW; P HST，in is the heat storage output of the heat storage system, kW; G HST，in is the heat storage capacity of the heat storage system, kW; P HTS，out is the heat release output of the heat storage system, kW; G HTS，out is the heat release capacity of the heat storage system, kW; P ICE is the output of the internal combustion engine, kW; G ICE is the capacity of the internal combustion engine, kW; P ASHP is the output of the electric refrigerator, kW; G ASHP is the capacity of the electric refrigerator, Kw; P GB is the output of the gas boiler, Kw; G GB is the capacity of the gas boiler, kW; P HP is the output of the geothermal heat pump, kW; G Hp is the capacity of the ground source heat pump, kW.

[0081] The output of the electric energy equipment is less than or equal to the equipment capacity constraint: P Battery，in ≤G Battery，in , PBattery，out ≤G Battery，out , P ASHP ≤G ASHP , P HP ≤G HP , where P represents the output of the equipment, kW; G represents the capacity of the equipment, kW; P Battery，in Indicates the charging output of the battery; G Battery，in is the charging capacity of the battery, kW; P Battery，out The discharge output of the battery; G Battery，out is the discharge capacity of the battery, kW; P ASHP is the output of the electric refrigerator, kW; G ASHP is the capacity of the electric refrigerator, Kw; P HP is the output of the geothermal heat pump, kW; G HP is the capacity of the ground source heat pump, kW.

[0082] For the construction of integrated energy systems, multi-timescale optimization control is performed based on deep reinforcement learning;

[0083] 1. Build a management model for the integrated energy system at the top, middle, and bottom levels, encompassing agents, environment spaces, action spaces, and reward functions. The action spaces for each layer represent the output of each energy supply device, while the environment spaces represent the system's cooling, heating, electricity, and gas loads, as well as the output of renewable energy, at each time period.

[0084] 2. Based on the deep reinforcement learning algorithm, interactive trial and error learning is carried out with the system environment to train the upper, middle and lower-level management models so that they can meet the real-time rolling management needs of the park's integrated energy system in offline mode.

[0085] (1) Agent action selection and state transition

[0086] At time slot t, the agent inputs the current environment state into the main policy network, which then outputs the action to be taken based on its deterministic behavioral policy. After selecting an action, the agent uses a decay coefficient to reduce the exploration factor to prevent overfitting.

[0087] (2) Gradient Agent Knowledge Storage

[0088] The agent performs a selected action in the environment, which then transitions to a new state and returns a corresponding reward. The agent integrates this state transition into a piece of "knowledge" and stores it in the experience replay pool. When the pool exceeds its capacity, the agent randomly extracts several pieces of "knowledge" at each iteration to help train the neural network.

[0089] (3) Agent network training

[0090] Combining the extracted "knowledge," the agent first passes the actions output by the main policy network and the current environment state to the main network to calculate their corresponding values ​​and loss function. Next, the agent calculates the gradient of its loss function with respect to the network parameters and uses the reinforcement learning optimizer to propagate the gradient information back and update the network. The agent then obtains the policy gradient of the policy by calculating the gradient of the discounted cumulative reward function with respect to the main policy network. Simultaneously, the agent combines the randomly extracted "knowledge" to calculate an unbiased estimate of the policy gradient and uses the reinforcement learning optimizer to propagate the policy gradient information back and update the network. Finally, the agent updates the target network and target policy network using a soft update method, using a soft update coefficient to control the weight of the original network parameters as new network parameters in each iterative update.

[0091] Finally, based on the optimized park-level integrated energy system structure and multi-time-scale optimization operation method, the system's economy (annualized cost), environmental protection (CO2 emissions), energy efficiency (primary energy utilization rate), and independence are calculated, and compared with the production-based system and the system with only a day-ahead scheduling plan. The advantages of the robust optimization and multi-time-scale optimization operation mode of the park-level integrated energy system are analyzed.

[0092] Example 2

[0093] The present invention provides a multi-time-scale optimization system for an integrated energy system, comprising:

[0094] Model building module: used to build a system multi-objective robust optimization configuration model based on renewable energy output and user load forecast results;

[0095] Computational solution module: used to solve the system's multi-objective robust optimization configuration model using a multi-objective genetic algorithm to determine the optimal system robust structural configuration;

[0096] Configuration optimization module: used to divide the integrated energy system into three time stages and optimize the three time stages according to the optimal system robust structure configuration.

[0097] Example 3

[0098] See also Figure 3 As shown, the present invention also provides an electronic device 100 for a multi-time-scale optimization method for an integrated energy system; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0099] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the integrated energy system multi-time-scale optimization method described in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area. The program storage area can store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data (such as audio data) created based on the use of the electronic device 100. In addition, the memory 101 can include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device.

[0100] The at least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100 and connects various parts of the entire electronic device 100 using various interfaces and lines.

[0101] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a multi-time-scale optimization method for an integrated energy system. The processor 102 can execute the plurality of instructions to implement:

[0102] Based on the forecast results of renewable energy output and user load, a multi-objective robust optimization configuration model for the system is constructed;

[0103] The multi-objective genetic algorithm is used to solve the system multi-objective robust optimization configuration model and determine the optimal system robust structure configuration;

[0104] The regulation process of the integrated energy system is divided into three time stages and optimized according to the optimal system robust structure configuration.

[0105] Example 4

[0106] If the module / unit integrated in the electronic device 100 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 present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0107] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0109] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0110] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A multi-time-scale optimization method for an integrated energy system, characterized by: include: Based on the forecast results of renewable energy output and user load, a multi-objective robust optimization configuration model for the system is constructed; The objective functions of the multi-objective robust optimization configuration model include: minimizing carbon emissions, minimizing annualized total costs, maximizing primary energy utilization, and maximizing the independence of the integrated energy system; the constraints of the multi-objective robust optimization configuration model include: energy supply and demand balance, and equipment capacity being less than or equal to the maximum capacity constraint of the equipment; The multi-objective genetic algorithm is used to solve the system multi-objective robust optimization configuration model and determine the optimal system robust structure configuration; The control process of the integrated energy system is divided into three time stages, and the three time stages are optimized according to the optimal system robust structure configuration; the three time stages include: a long time scale stage, a medium time scale stage, and a short time scale stage; The constraints in the long time scale stage and the medium time scale stage include: power balance constraint, heat balance constraint, cooling balance constraint, and equipment output being less than or equal to the maximum capacity constraint of the equipment; The constraints in the short time scale phase include: power balance constraint and power equipment output being less than or equal to the maximum capacity constraint of the equipment; The scheduling optimization objective in the long time scale stage is ,in, is the system operation cost, yuan; is the scheduling period, 24h; is the interaction cost between CCHP microgrid system and grid, RMB; is the natural gas fee, RMB; is the battery aging cost, yuan; is the operation and maintenance cost of the system, RMB; is the environmental cost of the system, yuan; The scheduling optimization objective of the medium time scale stage is: ,in, is the time span, 4h; is the regulation cost of grid interaction, RMB; is the fuel adjustment cost, yuan; Penalty cost for battery charge and discharge power variation, RMB; Penalty cost for heat storage / cold tank heat storage and release power change, RMB; The short-time-scale stage scheduling optimization objective is ,in, is the real-time power purchase adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time boiler power adjustment, kW; is the real-time electricity sales power adjustment, kW; is the real-time gas purchase power adjustment, kW; is the real-time power adjustment of the absorption chiller, kW; is the real-time power adjustment of the electric refrigerator, kW; is the real-time power adjustment of the ground source heat pump, kW.

2. The multi-time-scale optimization method for an integrated energy system according to claim 1, characterized in that: The energy supply and demand balance includes: electricity balance constraint, heat balance constraint and cooling balance constraint; Power balance constraints: ;in, It is hourly data; is the power generation of the gas turbine, kWh; is the amount of electricity purchased from the municipal power grid, kWh; is the photovoltaic power generation, kWh; is the battery discharge capacity, kWh; is the discharge capacity of wind power, kWh; is the park's electrical load, kWh; is the power consumption of the electric refrigerator, kWh; is the power consumption of the ground source heat pump, kWh; Heat balance constraints: ;in, is the heat release of the heat exchanger, kWh; Heat released to the boiler, kWh; The amount of heat collected by the solar thermal system, kWh; Heat released to the thermal storage system, kWh; is the park heat load, kWh; Cooling balance constraints: ,in, is the cooling capacity of the electric refrigerator, kWh; is the cooling capacity of the ground source heat pump, kWh; is the cooling capacity of the absorption refrigerant, kWh; is the cooling load of the park, kWh.

3. The multi-time-scale optimization method for an integrated energy system according to claim 1, characterized in that: The constraint that the device capacity is less than or equal to the device maximum capacity includes: , , , , , , , , , , , ,in, Indicates the capacity of the equipment, kW; is the charging capacity of the battery, kW; is the discharge capacity of the battery, kW; is the heat storage capacity of the heat storage system, kW; is the heat release capacity of the heat storage system, kW; is the capacity of the PV system, kW; is the capacity of the CSP system, kW; is the capacity of the internal combustion engine, kW; is the capacity of the electric refrigerator, kW; is the capacity of the gas boiler, kW; is the capacity of the wind turbine, kW; is the capacity of the ground source heat pump, kW; is the capacity of the absorption refrigerant, kW.

4. The multi-time-scale optimization method for an integrated energy system according to claim 1, characterized in that: The minimum carbon emissions are ,in, is the total carbon production during the entire life cycle of the park system operation. ; The carbon production during the entire life cycle of the park system in providing heating to users, ; The carbon production during the entire life cycle of the park system in the process of providing cooling to users, ; The carbon production during the entire life cycle of the park system in supplying power to users, .

5. The multi-time-scale optimization method for an integrated energy system according to claim 1, characterized in that: The minimum annualized total cost is , ,in, is the device category; is the annual total cost of the system, yuan; is the cost recovery factor; is the initial investment cost per unit capacity of the equipment, RMB / kW; is the equipment capacity, kW; is the number of devices; represents the interest rate; Indicates the life cycle of the equipment, in years; It is the equipment operation and maintenance cost coefficient, including system maintenance cost and fuel consumption cost.

6. The multi-time-scale optimization method for an integrated energy system according to claim 1, characterized in that: The maximum primary energy utilization rate is: ,in, The user's annual cooling, heating and electricity consumption, kWh; is the total energy consumed by the system, kWh.

7. The multi-time-scale optimization method for an integrated energy system according to claim 1, characterized in that: The highest level of independence of the integrated energy system is: ,in, It represents the total amount of electric energy purchased and sold by the system from the external large power grid and large thermal network, in kWh; It represents the total amount of thermal energy purchased and sold by the system from the external large power grid and large thermal network, in kWh; Refers to the total amount of cooling, heating and electricity load of the user, kWh.

8. The multi-time-scale optimization method for an integrated energy system according to claim 1, characterized in that: The device output is less than or equal to the device capacity constraint: , , , , , , , , ,in, Indicates the output of the equipment, kW; Indicates the capacity of the equipment, kW; Indicates the charging output of the battery; is the charging capacity of the battery, kW; Provide power for battery discharge; is the discharge capacity of the battery, kW; is the heat storage output of the heat storage system, kW; is the heat storage capacity of the heat storage system, kW; is the heat release output of the heat storage system, kW; is the heat release capacity of the heat storage system, kW; is the output of the internal combustion engine, kW; is the capacity of the internal combustion engine, kW; is the output of the electric refrigerator, kW; is the capacity of the electric refrigerator, Kw; is the output of the gas boiler, Kw; is the capacity of the gas boiler, kW; is the output of the geothermal heat pump, kW; is the capacity of the ground source heat pump, kW; The output of the electric energy equipment is less than or equal to the equipment capacity constraint: , , , ,in, Indicates the output of the equipment, kW; Indicates the capacity of the equipment, kW; Indicates the charging output of the battery; is the charging capacity of the battery, kW; Provide power for battery discharge; is the discharge capacity of the battery, kW; is the output of the electric refrigerator, kW; is the capacity of the electric refrigerator, Kw; is the output of the geothermal heat pump, kW; is the capacity of the ground source heat pump, kW.

9. Integrated energy system multi-time scale optimization system, characterized by: The multi-time-scale optimization method for an integrated energy system according to any one of claims 1 to 8 comprises: Model building module: used to build a system multi-objective robust optimization configuration model based on renewable energy output and user load forecast results; Computational solution module: used to solve the system's multi-objective robust optimization configuration model using a multi-objective genetic algorithm to determine the optimal system robust structural configuration; Configuration optimization module: used to divide the integrated energy system into three time stages and optimize the three time stages according to the optimal system robust structure configuration.

Images