A multi-scale optimization operation method for integrated energy systems considering dynamic characteristics

Abstract

The application belongs to the technical field of energy system management, and particularly relates to a multi-scale optimization operation method of an integrated energy system (IES) considering dynamic characteristics. The application mainly solves the problem that the dynamic characteristics of different energy devices in the IES are different, and the response rates of the devices to system load changes also have great differences, which has an impact on the actual operation of the integrated energy system and also significantly affects the overall comprehensive benefits of the system. The application proposes an IES multi-time scale optimization strategy considering the dynamic characteristics of the devices. The strategy includes day-ahead optimization, intra-day rolling optimization, climbing / descending optimization and real-time optimization, and there is a feedback process between each stage.

Description

Technical Field

[0001] This invention belongs to the field of energy system management technology, specifically relating to a multi-scale optimization operation method for a comprehensive energy system that considers dynamic characteristics. Background Technology

[0002] An integrated energy system refers to an integrated energy production, supply, and sales system formed through the organic coordination and optimization of energy generation, transmission and distribution, conversion, storage, and consumption during planning, construction, and operation. Therefore, integrated energy systems are hailed by the international energy community as the most likely mode of energy supply and utilization for human society 30-50 years from now. In the energy sector, there has long been a need for the coordinated optimization of different energy forms; almost every energy source requires the conversion and cooperation of multiple energy sources to achieve efficient utilization. Due to the differences in the development of different energy systems, energy supply is often planned, designed, and operated independently, lacking coordination. This results in problems such as low energy utilization rates, weak self-healing capabilities, and the need to improve the overall security of the energy supply system.

[0003] Integrated energy systems (IES) utilize the complementary characteristics of various energy types such as electricity, heat, and gas, as well as the principle of energy cascade utilization, to uniformly plan and coordinate the operation of multi-energy systems from four aspects: source, grid, load, and storage. This is conducive to promoting the consumption of renewable energy and is an important way to improve energy efficiency. However, the uncertainty of renewable energy output and system load poses challenges to the stable and reliable operation of the system.

[0004] To mitigate the adverse effects of source-load errors on system operation, scholars both domestically and internationally have conducted research on multi-timescale optimization of energy systems (IES). Some scholars have proposed and established a two-stage multi-timescale model predictive control scheduling strategy, incorporating rolling optimization and dynamic adjustment stages, to enhance system scheduling flexibility. They have also proposed a probabilistic energy management method that considers both day-ahead and real-time scheduling. Other scholars have proposed a multi-timescale rolling optimization operation method, developing unit output plans in two stages: "day-ahead" and "intra-day," to reduce the impact of source-load fluctuations on the actual operation of IES. Furthermore, a multi-timescale low-carbon operation optimization strategy considering electricity-gas-heat-hydrogen demand response and a tiered carbon emission cost mechanism has been proposed, establishing a three-stage multi-timescale optimization model: "day-ahead," "intra-day," and "real-time." However, the dynamic characteristics of different energy devices within the IES differ, and the rates at which each device responds to system load changes vary significantly, which can substantially affect the overall comprehensive benefits of the system. Simply introducing equipment ramp power constraints to reflect equipment response capabilities is insufficient to fully reflect the impact of different dynamic characteristics of equipment on the actual operation of the system. Therefore, it is necessary to fully consider the dynamic operating characteristics of the equipment and formulate detailed day-day-intraday-real-time optimized operation plans for the integrated energy system.

[0005] To address this, this method proposes an IES multi-timescale optimization strategy that considers the dynamic characteristics of equipment. Optimized scheduling models are constructed at three timescales: day-ahead, intraday, and real-time. Furthermore, an optimization model for ramping and descent of "slow-response" equipment is established based on the dynamic characteristics of the equipment. Summary of the Invention

[0006] To address the technical problems existing in the prior art, this invention proposes a multi-scale optimization operation method for integrated energy systems that considers dynamic characteristics, including day-ahead optimization, intraday rolling optimization, ramp / descent optimization, and real-time optimization, with feedback processes existing between each stage.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0008] This invention provides a multi-scale optimization operation method for integrated energy systems that considers dynamic characteristics, including:

[0009] (1) Day-ahead optimization: Based on the day-ahead renewable energy output and user load forecast data, and considering the time-of-use electricity price and the equipment operation constraints in the system, with the goal of minimizing daily operating costs and a time scale of 1 hour, a day-ahead scheduling plan is formulated. The cooling, heating and power inequality equation is established through the load margin, and on this basis, an adjustment benchmark is provided for intraday rolling optimization and real-time optimization.

[0010] (2) Intraday rolling optimization: Renewable energy output and user load data are predicted and updated using 15-minute time scale; the daily scheduling plan within the calculation time domain is modified using 2-hour rolling calculation time domain. The optimization objective is to minimize the energy purchase cost and equipment output adjustment penalty cost within the rolling calculation time domain, and obtain the intraday scheduling plan within the calculation time domain.

[0011] (3) Climbing / Descending Optimization: Nested with real-time optimization and intraday rolling optimization, optimization calculations are performed with a 5-minute time scale to correct the intraday scheduling plan in the current calculation time domain; within the calculation time domain of the intraday scheduling plan, considering the dynamic characteristics of slow-response equipment, climbing / descending optimization is performed using a genetic algorithm;

[0012] (4) Real-time optimization: Based on real-time forecast data and slow-response equipment ramp-up plans, the output of equipment other than slow-response equipment in the daily scheduling plan is adjusted in real time. The optimization goal is to minimize the system's cold and hot load deficit penalties and equipment power adjustment penalties within the calculation time domain.

[0013] Furthermore, there is a feedback process among the four stages of day-ahead optimization, intraday rolling optimization, ramp / descend optimization, and real-time optimization. Specifically, after intraday optimization, ramp / descend optimization, and real-time optimization, the optimal real-time scheduling plan for each device in this calculation time domain is obtained; the optimal real-time scheduling plan for the first 15 minutes is executed, while the calculation time domain is rolled forward by 15 minutes; the energy storage state of the energy storage device at the 15th minute is used as the initial constraint condition to start the intraday optimization calculation for the next calculation time domain, and the previous optimization process is repeated continuously.

[0014] Furthermore, in an integrated energy system, in addition to the constraints of each device, the following constraints also need to be met:

[0015] Electrical balance constraints:

[0016] E PV (t)+E ICE (t)+E bat,out (t)+E grid,buy (t)

[0017] =L E (t)+E EC (t)+E bat,in (t)+E grid,sell (t)

[0018] In the formula: L E (t) represents the electrical load of the integrated energy system during time period t; E ICE (t) represents the power generation of the gas-fired internal combustion engine during time period t; E PV (t) represents the power generation of the photovoltaic panel during time period t; E EC(t) represents the power consumption of the electric chiller during time period t; E bat,in (t) and E bat,out (t) represent the battery charging power and discharging power during time period t, respectively; E grid,buy (t) and E grid,sell (t) represents the electricity purchased and sold by the integrated energy system during time period t;

[0019] Thermal equilibrium constraint:

[0020] Q WB (t)+Q GB (t)+Q ST (t)+Q HST,out (t)

[0021] =L Q (t) / η HE +Q HST,in (t)+Q AC (t)

[0022] In the formula: Q GB (t) represents the upper limit of the ramp rate of the absorption chiller during time period t; Q WB (t) represents the amount of waste heat recovered by the waste heat boiler during time period t; Q ST (t) represents the collector power of the solar collector during time period t; Q HST,in (t) and Q HST,out (t) represents the heat storage and heat release power of the hot water tank during time period t; Q AC (t) represents the heat consumption of the absorption chiller during time period t; L Q (t) represents the heat load of the integrated energy system during time period t; η HE The heat transfer coefficient of the heat exchanger;

[0023] Cold balance constraint:

[0024] C AC (t)+C EC (t)+C CST,out (t)=L C (t)+C CST,in (t)

[0025] In the formula: L C (t) represents the cooling load of the system during time period t; C EC (t) represents the cooling power of the electric refrigeration unit during time period t; C CST,in (t) and C CST,out (t) represents the cooling storage and cooling discharge power of the cold water tank during time period t; C AC (t) represents the cooling power of the absorption chiller during time period t;

[0026] Energy storage device start-stop state constraints:

[0027] Wi (T rq ) = W i (1)

[0028] In the formula: T rq To optimize the computation time domain; W i Store energy for the i-th type of energy storage device.

[0029] Furthermore, by establishing the cooling-heating-electricity inequality equation through the load margin, it can be expressed as:

[0030] E all (t)∈[L E (t)·(1-σ),L E (t)·(1+σ)]

[0031] Q all (t)∈[Q E (t)·(1-σ),Q E (t)·(1+σ)]

[0032] C all (t)∈[C E (t)·(1-σ),C E (t)·(1+σ)]

[0033] In the formula: E all Q all C all These represent the total power output for power supply, heating, and cooling in the integrated energy system; σ is the margin coefficient.

[0034] Furthermore, daily operating costs include equipment operating costs, gas purchase costs, grid interaction costs, and battery aging costs, expressed as:

[0035]

[0036] In the formula: P om For equipment operating costs; P gas For the system's gas purchase cost; P grid For grid interaction costs; P bat Costs associated with battery aging;

[0037] Among them, equipment operating cost P om Represented as:

[0038] P om (t)=[E PV (t)·P om,PV +Q ST (t)·P om,ST +E ICE (t)·P om,ICE

[0039] +(E bat,in (t)+E bat,out (t))·P om,bat +L Q (t) / η HE ·P om,HE

[0040] +(Q HST,in (t)+Q HST,out (t))·P om,HST +Q WB (t)·P om,WB

[0041] +(C CST,in (t)+C CST,out (t))·P om,CST +C AC (t)·P om,AC

[0042] +Q GB (t)·P om,GB +C EC (t)·P om,EC ]·Δt

[0043] In the formula: P om,PV PV operating cost; P om,ICE For ICE operating costs; P om,bat For the operating cost of BAT (bat) systems; P om,HE For HE operating costs; P om,ST For ST operating costs; P om,GB GB operating cost; P om,WB For WB operating costs; P om,HST For HST operating costs; P om,EC For EC operating costs; P om,AC For AC operating costs; P om,CST CST operating costs; L Q (t) represents the heat load of the integrated energy system during time period t; η HE The heat transfer coefficient of the heat exchanger;

[0044] Gas purchase cost P gas Represented as:

[0045]

[0046] In the formula: R gas For natural gas prices; H gas The lower heating value of natural gas; F ICE (t) represents the natural gas power consumed by the gas-fired internal combustion engine during time period t; F GB (t) represents the natural gas power consumed by the gas-fired boiler during time period t;

[0047] Grid interaction cost P grid Represented as:

[0048] C grid (t)=[E grid,buy (t)·P grid,buy (t)

[0049] -E grid,sell (t)·P grid,sell (t)]·Δt

[0050] In the formula: P grid,buy (t) and P grid,sell (t) represents the purchase and sale prices of electricity during time period t, respectively;

[0051] Battery aging cost P bat Represented as:

[0052] P bat (t)=[U bat,in (t)+U bat,out (t)]·R bat

[0053] In the formula: R bat Cost of battery charge / discharge conversion; U bat,in (t) and U bat,out (t) represents the battery charging and discharging start flags at time t; flag bat,in and flag bat,out These represent the battery charging and discharging indicators at time t; U bat,in (t) and U bat,out (t) respectively satisfy the battery charge / discharge state transition constraints:

[0054]

[0055] Furthermore, the objective function for intraday rolling optimization is:

[0056]

[0057]

[0058] In the formula: P dom Adjusting the penalty cost for equipment output; the superscript rq indicates the planned output of each piece of equipment for the day, and rn indicates the optimized output of each piece of equipment within the day; C AC For the output of the absorption chiller; C EC Power output for electric chillers; E ICE Power is supplied to the gas-fired internal combustion engine; Q GB Power output for gas-fired boilers; Q HST,in and Q HST,out For the heat storage and release power of the hot water tank; CCST,in and C CST,out E is used to store cooling capacity in a cold water storage tank. bat,in and E bat,out Battery charging and discharging power; E grid,buy and E grid,sell Power purchased and sold to the power grid; μ dom,AC Penalty factor for power adjustment of absorption chiller; μ dom,EC Penalty factor for adjusting the power of electric chillers; μ dom,ICE Penalty factor for adjusting the power of gas internal combustion engines; μ dom,GB Penalty factor for adjusting the power output of gas-fired boilers; μ dom,HST Adjust the penalty coefficient for the heat storage and release power of the hot water storage tank; μ dom,CST Adjust the penalty coefficient for the cold storage capacity of the cold water tank; μ dom,bat Adjust the penalty coefficient for battery charging and discharging power; μ dom,grid The penalty factor for adjusting the power exchange between the power grids.

[0059] Furthermore, slow-response devices include gas-fired internal combustion engines, gas-fired boilers, and absorption chillers.

[0060] Furthermore, a genetic algorithm is used for hill climbing / descent optimization, specifically as follows:

[0061] Optimization variable: The ramp / deceleration plan of slow-response equipment. The ramp / deceleration plan of slow-response equipment is composed of the ramp / deceleration modes adopted by the equipment in various time periods. Each equipment adjusts its power according to two modes in each time period within the calculation time domain:

[0062] Mode 1: Start adjusting at the beginning of the time period until the planned output for the next time period is reached;

[0063] Mode 2: Start adjusting during the time period and reach the planned output for the next time period at the end of the time period.

[0064] Furthermore, the system's cooling and heating load deficit penalties and equipment power adjustment penalties are minimized, as expressed as:

[0065]

[0066] In the formula: α1 and α2 are the cold / heat deficit penalty coefficient and the equipment power adjustment penalty coefficient, respectively; Q lack and C lack These are respectively a lack of heat and a lack of cold; P ad This is the sum of squares of the relative power adjustments of each device;

[0067] Calculation of heat and cold deficits:

[0068]

[0069] Calculation of relative power adjustment of equipment

[0070]

[0071] In the formula: the superscript rn represents the daily optimized output of each device, ss represents the real-time optimized output of the device; the subscript max represents the upper limit of the device output; E bat,in and E bat,out Battery charging and discharging power; E grid,buy and E grid,sell C. Purchase and sell electrical power to the power grid; EC It provides power to the electric refrigeration unit.

[0072] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0073] This invention proposes a multi-scale optimization operation method for integrated energy systems that considers dynamic characteristics. It fully considers the dynamic operating characteristics of equipment and refines the day-ahead, intraday, and real-time optimization operation schemes for the integrated energy system. Optimization scheduling models are constructed at the day-ahead, intraday, and real-time time scales, and an optimization model for the ramp-up and descent of "slow-response" equipment is established based on the dynamic characteristics of the equipment. This improves the overall comprehensive efficiency of the integrated energy system. Attached Figure Description

[0074] Figure 1 A diagram of the integrated energy system structure;

[0075] Figure 2 To optimize the calculation of the time period relationship diagram;

[0076] Figure 3 To optimize the calculation flowchart;

[0077] Figure 4 These are schematic diagrams illustrating two climbing modes;

[0078] Figure 5 The diagram shows the equipment output under different optimization strategies, where (a) is a typical daytime gas-fired internal combustion engine in winter, (b) is a typical daytime gas-fired boiler in winter, (c) is a typical daytime gas-fired internal combustion engine in summer, and (d) is a typical daytime absorption chiller in summer. Detailed Implementation

[0079] To facilitate understanding of the present invention, a more comprehensive description will be given below. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the present invention.

[0080] Example 1

[0081] like Figure 1As shown, the integrated energy system in this embodiment mainly includes photovoltaic power generation equipment, solar thermal collectors, gas internal combustion engines, waste heat boilers, gas boilers, heat exchangers, electric chillers, absorption chillers, batteries, hot water storage tanks, and cold water storage tanks, and can also exchange electrical energy with the external power grid.

[0082] The system's electrical load is supplied by photovoltaic power generation equipment, a gas-fired internal combustion engine, and the external power grid; the heat load is supplied by a gas-fired internal combustion engine, a gas-fired boiler, and solar thermal collectors; and the cooling load is supplied by an electric chiller and an absorption chiller. Waste heat from the gas-fired internal combustion engine is recovered and utilized by a waste heat boiler; the low-grade heat stored in the hot water storage tank is not used to drive the absorption chiller.

[0083] 1. Gas internal combustion engine

[0084] A gas-fired internal combustion engine consumes natural gas to power and heat the system; its natural gas power consumption can be expressed as:

[0085]

[0086] In the formula: E ICE (t) represents the power generation of the gas-fired internal combustion engine during time period t, in kWh; F ICE (t) represents the natural gas power consumed by the gas-fired internal combustion engine during time period t, in kW; η ICE The heat-to-work conversion efficiency of the gas internal combustion engine is taken as 0.25; η m The mechanical efficiency of the gas internal combustion engine is taken as 0.93.

[0087] The starting power and ramping power constraints of a gas-fired internal combustion engine can be expressed as:

[0088] flag ICE (t)·E ICE,min ≤E ICE (t)≤flag ICE (t)·E ICE,max

[0089] -E ICE,dmax ·Δt≤E ICE (t)-E ICE (t-1)≤E ICE,dmax ·Δt

[0090] In the formula: E ICE，max and E ICE，min These are the maximum and minimum operating power of the gas-fired internal combustion engine, respectively, in kW; flag ICE E is the start sign for a gas-fired internal combustion engine. ICE,dmax Δt represents the upper limit of the electric power ramp rate of the gas-fired internal combustion engine, in kW / h; Δt is the calculation time step, in h.

[0091] The minimum start-stop time constraint for a gas-fired internal combustion engine can be expressed as:

[0092]

[0093]

[0094] In the formula: t ICE,on and t ICE,off These are the shortest start-up and shutdown times for a gas-fired internal combustion engine, respectively, in hours (h).

[0095] 2. Waste heat boiler

[0096] A waste heat boiler recovers waste heat from a gas-fired internal combustion engine to power the system; its output thermal power can be expressed as:

[0097]

[0098] In the formula: Q WB (t) represents the waste heat recovery rate of the waste heat boiler during time period t, in kW; η loss η is the heat dissipation loss rate of the gas internal combustion engine, taken as 0.3; WB The waste heat recovery efficiency is set to 0.8.

[0099] 3. Gas-fired boiler

[0100] The heating power of a gas-fired boiler can be expressed as:

[0101] Q GB (t)=F GB (t)·η GB

[0102] In the formula: Q GB (t) represents the output thermal power of the gas-fired boiler during time period t, in kW; F GB (t) represents the natural gas power consumed by the gas-fired boiler during time period t, in kW; η GB This refers to the heating efficiency of a gas-fired boiler.

[0103] The ramp-up power constraint for gas-fired boilers can be expressed as:

[0104] -Q GB,dmax ·Δt≤Q GB (t)-Q GB (t-1)≤Q GB,dmax ·Δt

[0105] In the formula: Q WB,dmax This represents the upper limit of the ramp rate for absorption chillers, expressed in kW / h.

[0106] 4. Photovoltaic power generation equipment

[0107] The power generation calculation model for photovoltaic power generation equipment is as follows:

[0108] E PV (t)=E PV,0 (t)·f t (t)·f s ·f μ ·f o

[0109]

[0110]

[0111]

[0112] f t (t)=1-0.005×(T) PV (t)-25)

[0113]

[0114] In the formula: E PV,0 The theoretical power generation of the photovoltaic panel is expressed in kW; f t Temperature influence factor; f s The dust accumulation factor is set to 0.98; f μ f is the performance mismatch factor, taken as 0.95; o The influence factor for other factors is set at 0.98; I PV I represents the operating current of the photovoltaic panel, measured in amperes (A). sc0 The short-circuit current of the photovoltaic panel is taken as 7.91A; I pm0 The peak current of the photovoltaic panel is taken as 7.45A; Solar irradiance during time period t, in W / m² 2 ; The solar irradiance of the photovoltaic panel under standard conditions is taken as 1000 W / m. 2 V PV This refers to the operating voltage of the photovoltaic panel, measured in V. pm0 T represents the peak voltage of the photovoltaic panel, taken as 36.1V. PV T represents the actual surface temperature of the photovoltaic panel, in °C. amb The ambient temperature is expressed in °C.

[0115] 5. Solar thermal collector

[0116] The heat collection power of a solar thermal collector can be expressed as:

[0117]

[0118]

[0119] In the formula: Q STThe solar collector's heat collection power is expressed in kW; S ST The effective heat collection area of ​​the solar collector is expressed in m². 2 ;X ST X represents the heat collection efficiency of the solar collector. ST,0 U is the intercept of the instantaneous heat collection efficiency of the solar collector, taken as 0.7; ST The heat loss coefficient of the solar collector is taken as 2.5 W / (m²). 2 ·K); T w,in The inlet water temperature of the solar collector is set to 25℃.

[0120] 6. Electric refrigeration unit

[0121] The cooling power of an electric chiller can be expressed as:

[0122] C EC (t)=COP EC ·E EC (t)

[0123] In the formula: C EC Cooling power of the electric refrigeration unit, in kW; COP EC E is the coefficient of performance for electrical cooling. EC The power consumption of the electric chiller is expressed in kW.

[0124] The ramp-up power constraint of an electric chiller can be expressed as:

[0125] -C EC,dmax ·Δt≤C EC (t)-C EC (t-1)≤C EC,dmax ·Δt

[0126] In the formula: C EC,dmax This represents the upper limit of the ramp rate for electric chillers, expressed in kW / h.

[0127] 7. Absorption chiller

[0128] The cooling capacity of an absorption chiller can be expressed as:

[0129] C AC (t)=COP AC ·Q AC (t)

[0130] In the formula: C AC This refers to the refrigeration power of an absorption refrigeration system, measured in kW; COP. AC Q is the coefficient of performance for absorption refrigeration. AC This represents the heat consumption of an absorption chiller, expressed in kW.

[0131] The ramp-up power constraint of an absorption chiller can be expressed as:

[0132] -CAC,dmax ·Δt≤C AC (t)-C AC (t-1)≤C AC,dmax ·Δt

[0133] In the formula, C AC,dmax This represents the upper limit of the ramp rate for absorption chillers, expressed in kW / h.

[0134] 8. Storage battery

[0135] The constraints on battery storage capacity and charge / discharge power can be expressed as:

[0136]

[0137] E bat,min ≤E bat (t)≤E bat,max

[0138]

[0139] In the formula: E bat The amount of energy stored in a battery is expressed in kW·h; r bat E represents the self-discharge rate, taken as 0.001. bat,max and E bat,min These represent the maximum and minimum battery capacity, respectively, in kWh; E bat,in and E bat,out These represent the charging and discharging power of the battery, respectively, in kW; E bat,in,max and E bat,out,max These represent the upper limits of battery charging and discharging power, respectively, in kW; η bat,in and η bat,out These represent the battery charging efficiency and discharging efficiency, both taken as 0.9; flag bat,in and flag bat,out These are the battery charging and discharging indicators.

[0140] The battery charging and discharging power ramp-up constraint can be expressed as:

[0141]

[0142] In the formula: E bat,in,dmax and E bat,out,dmin These represent the upper limits of the battery's charge / discharge ramp rate, in kW / h.

[0143] 9. The heat storage capacity and heat release power constraints of the hot water storage tank can be expressed as:

[0144]

[0145] Q HST,min ≤QHST (t)≤Q HST,max

[0146]

[0147] In the formula: Q HST The heat storage capacity of the hot water tank is expressed in kWh; r HST The heat loss rate is taken as 0.005; Q HST,in and Q HST,out These represent the heat storage and heat release power of the hot water storage tank, respectively, in kW; Q HST,max and Q HST,min Q represents the maximum and minimum heat storage capacity of the hot water storage tank, respectively, in kWh; HST,in,max and Q HST,out,max These represent the upper limits of the heat storage and heat release power of the hot water tank, respectively, in kW; η HST,in and η HST,out The values ​​represent the heat storage and heat release efficiencies of the hot water storage tank, both taken as 0.9; flag HST,in and flag HST,out These are the heat storage and release indicators for hot water storage tanks.

[0148] 10. Cold water storage tank

[0149] The constraints on the cold storage capacity and cold storage / release power of the cold water storage tank can be expressed as:

[0150]

[0151] C CST,min ≤C CST (t)≤C CST,max

[0152]

[0153] In the formula: C CST The cold storage capacity of the cold water tank is expressed in kWh; r CST The cold loss rate is set to 0.005; C CST,in and C CST,out These represent the cooling storage and cooling discharge capacities of the cold water tank, respectively, in kW; C CST,max and C CST,min These represent the maximum and minimum cooling capacity of the cold water storage tank, respectively, in kWh; C CST,in,max and C CST,out,max These represent the upper limits of the cold storage and cooling power of the cold water tank, respectively, in kW; η CST,in and η CST,out The values ​​represent the cold storage and cooling efficiency of the cold water tank, both taken as 0.9; flag CST,in and flag CST,out These are the markings for storing and releasing cold water in the cold water storage tank.

[0154] 11. External power grid

[0155] The power constraint between the system and the external power grid can be expressed as:

[0156]

[0157] In the formula: E grid,buy (t) and E grid,sell (t) represents the system's purchased and sold electricity power during time period t, respectively, in kW; E grid,buy,max (t) and E grid,sell,max (t) represents the upper limit of the system's power purchase and sales capacity during time period t, in kW; flag grid,buy (t) and flag grid,sell (t) represents the electricity purchase and sale indicators for time period t.

[0158] A multi-scale optimization operation method for integrated energy systems considering dynamic characteristics includes day-ahead optimization, intraday rolling optimization, ramp / descent optimization, and real-time optimization, with feedback processes existing between each stage. The computation time periods of each stage are as follows: Figure 2 As shown, the optimized calculation process is as follows: Figure 3 As shown. Day-ahead optimization: Based on day-ahead renewable energy output and user load forecast data, and considering time-of-use pricing and equipment operation constraints within the system, a day-ahead scheduling plan is formulated with the goal of minimizing daily operating costs and a time scale of 1 hour. A cooling-heating-power inequality equation is established through a certain load margin, and an adjustment benchmark is provided for intraday rolling optimization and real-time optimization based on this.

[0159] By establishing the cooling-heating-electricity inequality equation through the load margin, it can be expressed as:

[0160] E all (t)∈[L E (t)·(1-σ),L E (t)·(1+σ)]

[0161] Q all (t)∈[Q E (t)·(1-σ),Q E (t)·(1+σ)]

[0162] C all (t)∈[C E (t)·(1-σ),C E (t)·(1+σ)]

[0163] In the formula: E all Q all C all These represent the total power output for power supply, heating, and cooling in the integrated energy system; σ is the margin coefficient.

[0164] Daily operating costs include equipment operating costs, gas purchase costs, grid connection costs, and battery aging costs, expressed as:

[0165]

[0166] In the formula: P om For equipment operating costs; P gas For the system's gas purchase cost; P grid For grid interaction costs; P bat Costs associated with battery aging;

[0167] Among them, equipment operating cost P om Represented as:

[0168] P om (t)=[E PV (t)·P om,PV +Q ST (t)·P om,ST +E ICE (t)·P om,ICE

[0169] +(E bat,in (t)+E bat,out (t))·P om,bat +L Q (t) / η HE ·P om,HE

[0170] +(Q HST,in (t)+Q HST,out (t))·P om,HST +Q WB (t)·P om,WB

[0171] +(C CST,in (t)+C CST,out (t))·P om,CST +C AC (t)·P om,AC

[0172] +Q GB (t)·P om,GB +C EC (t)·P om,EC ]·Δt

[0173] In the formula: P om,PV PV operating cost; P om,ICE For ICE operating costs; P om,bat For the operating cost of BAT (bat) systems; P om,HE For HE operating costs; P om,STFor ST operating costs; P om,GB GB operating cost; P om,WB For WB operating costs; P om,HST For HST operating costs; P om,EC For EC operating costs; P om,AC For AC operating costs; P om,CST CST operating costs; L Q (t) represents the heat load of the integrated energy system during time period t; η HE The heat transfer coefficient of the heat exchanger;

[0174] Gas purchase cost P gas Represented as:

[0175]

[0176] In the formula: R gas For natural gas prices; H gas The lower heating value of natural gas; F ICE (t) represents the natural gas power consumed by the gas-fired internal combustion engine during time period t; F GB (t) represents the natural gas power consumed by the gas boiler during time period t; Δt is the calculation time step;

[0177] Grid interaction cost P grid Represented as:

[0178] C grid (t)=[E grid,buy (t)·P grid,buy (t)

[0179] -E grid,sell (t)·P grid,sell (t)]·Δt

[0180] In the formula: P grid,buy (t) and P grid,sell (t) represents the purchase and sale prices of electricity during time period t, respectively;

[0181] Battery aging cost P bat Represented as:

[0182] P bat (t)=[U bat,in (t)+U bat,out (t)]·R bat

[0183] In the formula: R bat Cost of battery charge / discharge conversion; U bat,in (t) and U bat,out (t) represents the battery charging and discharging start flags at time t; flag bat,in and flag bat,outThese represent the battery charging and discharging indicators at time t; U bat,in (t) and U bat,out (t) respectively satisfy the battery charge / discharge state transition constraints:

[0184]

[0185] Intraday rolling optimization: Renewable energy output and user load data are predicted and updated using a 15-minute time scale; the day-ahead scheduling plan within the calculation time domain is revised using a 2-hour rolling calculation time domain. The optimization objective is to minimize the energy purchase cost and equipment output adjustment penalty cost within the rolling calculation time domain, thereby obtaining the intraday scheduling plan within the calculation time domain.

[0186] The objective function for intraday rolling optimization is:

[0187]

[0188]

[0189] In the formula: P dom Adjusting the penalty cost for equipment output; the superscript rq indicates the planned output of each piece of equipment for the day, and rn indicates the optimized output of each piece of equipment within the day; C AC For the output of the absorption chiller; C EC Power output for electric chillers; E ICE Power is supplied to the gas-fired internal combustion engine; Q GB Power output for gas-fired boilers; Q HST,in and Q HST,out For the heat storage and release power of the hot water tank; C CST,in and C CST,out E is used to store cooling capacity in a cold water storage tank. bat,in and E bat,out Battery charging and discharging power; E grid,buy and E grid,sell Power purchased and sold to the power grid; μ dom,AC Penalty factor for power adjustment of absorption chiller; μ dom,EC Penalty factor for adjusting the power of electric chillers; μ dom,ICE Penalty factor for adjusting the power of gas internal combustion engines; μ dom,GB Penalty factor for adjusting the power output of gas-fired boilers; μ dom,HST Adjust the penalty coefficient for the heat storage and release power of the hot water storage tank; μ dom,CST Adjust the penalty coefficient for the cold storage capacity of the cold water tank; μ dom,bat Adjust the penalty coefficient for battery charging and discharging power; μ dom,grid The penalty factor for adjusting the power exchange between the power grids.

[0190] Climbing / Descending Optimization: Nested with real-time optimization and intraday rolling optimization, it performs optimization calculations on a 5-minute time scale to revise the intraday scheduling plan in the current calculation time domain; within the calculation time domain of the intraday scheduling plan, considering the dynamic characteristics of slow-response devices, it uses a genetic algorithm for climbing / descending optimization.

[0191] Optimization variable: The ramp / deceleration plan of slow-response equipment. The ramp / deceleration plan of slow-response equipment is composed of the ramp / deceleration modes adopted by the equipment in various time periods. Each equipment adjusts its power according to two modes in each time period within the calculation time domain:

[0192] Mode 1: Start adjusting at the beginning of the time period until the planned output for the next time period is reached;

[0193] Mode 2: Start adjusting during the time period and reach the planned output for the next time period at the end of the time period.

[0194] Taking hill climbing as an example, the two modes are as follows: Figure 4 As shown. Where, P m (t) represents the planned daily output of a certain device during time period t. The real-time output plan of the device is determined based on the ramping mode, the daily output plan, and the device ramping speed, and is used as the initial condition for real-time optimization.

[0195] Real-time optimization: Based on real-time forecast data and slow-response equipment ramp-up plans, the output of equipment other than slow-response equipment in the daily scheduling plan is adjusted in real time. The optimization goal is to minimize the system's cold and hot load deficit penalties and equipment power adjustment penalties within the calculation time domain.

[0196] In the real-time optimization phase, only the output of electrically driven "fast-response" equipment is adjusted. The output of other equipment follows the intraday optimization and ramp / deceleration scheduling plans. During the optimization process, cold and heat balance constraints are no longer required. The optimization objective in the real-time optimization phase is to minimize the system's cold and heat load deficit penalties and equipment power adjustment penalties, which can be expressed as:

[0197]

[0198] In the formula: α1 and α2 are the cold / heat deficit penalty coefficient and the equipment power adjustment penalty coefficient, respectively; Q lack and C lack These are respectively a lack of heat and a lack of cold; P ad This is the sum of squares of the relative power adjustments of each device;

[0199] Calculation of heat and cold deficits:

[0200]

[0201] Calculation of relative power adjustment for equipment:

[0202]

[0203] In the formula: the superscript rn represents the daily optimized output of each device, ss represents the real-time optimized output of the device; the subscript max represents the upper limit of the device output; E bat,in and E bat,out Battery charging and discharging power; E grid,buy and E grid,sell C. Purchase and sell electrical power to the power grid; EC It provides power to the electric refrigeration unit.

[0204] Feedback Process: There is a feedback process between the four stages of day-ahead optimization, intraday rolling optimization, ramp / decline optimization, and real-time optimization. After intraday optimization, ramp / decline optimization, and real-time optimization, the optimal real-time scheduling plan for each device in this calculation time domain is obtained; the optimal real-time plan for the first 15 minutes is executed, and the calculation time domain is rolled forward by 15 minutes; the energy storage status of the energy storage device at the 15th minute is used as the initial constraint condition to start the intraday optimization calculation for the next calculation time domain, and the previous optimization process is repeated continuously.

[0205] In the multi-timescale optimization calculation process, in the integrated energy system, in addition to the constraints of each device, the following constraints also need to be met:

[0206] Electrical balance constraints:

[0207] E PV (t)+E ICE (t)+E bat,out (t)+E grid,buy (t)

[0208] =L E (t)+E EC (t)+E bat,in (t)+E grid,sell (t)

[0209] In the formula: L E (t) represents the electrical load of the integrated energy system during time period t; E ICE (t) represents the power generation of the gas-fired internal combustion engine during time period t; E PV (t) represents the theoretical power generation of the photovoltaic panel during time period t; E EC (t) represents the power consumption of the electric chiller during time period t; E bat,in (t) and E bat,out (t) represent the battery charging power and discharging power during time period t, respectively; E grid,buy (t) and E grid,sell (t) represents the electricity purchased and sold by the integrated energy system during time period t;

[0210] Thermal equilibrium constraint:

[0211] Q WB (t)+Q GB (t)+Q ST (t)+Q HST,out (t)

[0212] =L Q (t) / η HE +Q HST,in (t)+Q AC (t)

[0213] In the formula: Q GB (t) represents the upper limit of the ramp rate of the absorption chiller during time period t; Q WB (t) represents the amount of waste heat recovered by the waste heat boiler during time period t; Q ST (t) represents the collector power of the solar collector during time period t; Q HST,in (t) and Q HST,out (t) represents the heat storage and heat release power of the hot water tank during time period t; Q AC (t) represents the heat consumption of the absorption chiller during time period t; L Q (t) represents the heat load of the integrated energy system during time period t; η HE The heat transfer coefficient of the heat exchanger;

[0214] Cold balance constraint:

[0215] C AC (t)+C EC (t)+C CST,out (t)=L C (t)+C CST,in (t)

[0216] In the formula: L C (t) represents the cooling load of the system during time period t; C EC (t) represents the cooling power of the electric refrigeration unit during time period t; C CST,in (t) and C CST,out (t) represents the cooling storage and cooling discharge power of the cold water tank during time period t; C AC (t) represents the cooling power of the absorption chiller during time period t;

[0217] Energy storage device start-stop state constraints:

[0218] W i (T rq ) = W i (1)

[0219] In the formula: T rq To optimize the computation time domain; W i Store energy for the i-th type of energy storage device.

[0220] Example 2

[0221] Analysis of Climbing / Descent Optimization Results

[0222] To demonstrate the characteristics of multi-timescale optimization scheduling that considers equipment dynamics, the optimization results are compared and analyzed with those that do not consider equipment dynamics. The multi-timescale optimization strategy used for comparison does not perform ramp / descent optimization, and the optimization methods used in the day-ahead, intra-day, and real-time optimization stages are the same as those of the proposed strategy.

[0223] Real-time output of a typical winter gas-fired internal combustion engine, a typical winter gas-fired boiler, a typical summer gas-fired internal combustion engine, and a typical summer absorption chiller during certain periods under two strategies are as follows: Figure 5 As shown in (a), (b), (c), and (d), the selected time periods are mainly daytime, when the output of various loads and equipment is at a relatively high level. It can be seen that under both strategies, the real-time output of the three types of equipment fluctuates around the planned output for the previous day. When load fluctuations occur, priority is given to adjusting the real-time output of the gas-fired internal combustion engine, which fluctuates more significantly relative to the planned output for the previous day; the real-time output of the gas-fired boiler and absorption chiller is closer to the planned output for the previous day. Under the strategy that does not consider the dynamic characteristics of the equipment, the dynamic response start time of each piece of equipment is relatively late, and the equipment output exhibits a sudden rise and fall during the process of meeting the scheduling requirements; under the strategy that considers the dynamic characteristics of the equipment, each piece of equipment starts responding earlier, and the dynamic process curve of the equipment's real-time output slows down significantly.

[0224] The above results show that optimizing by considering the dynamic characteristics of the equipment during the multi-timescale optimization process is more in line with the actual operating results and can effectively reduce the fluctuation of real-time output of the equipment, thus achieving economical and stable operation of the equipment.

[0225] (2) Analysis of optimization results at multiple time scales

[0226] The results of multi-timescale optimization are compared with those of day-ahead optimization only. When the system performs day-ahead optimization only, the actual operation follows the day-ahead scheduling plan, and the power deficit caused by fluctuations in electricity, heat, and cooling loads is supplied by the external power grid, gas boilers, and electric chillers, respectively. The grid interaction power and daily operating costs for each typical day with and without multi-timescale optimization are shown in Tables 1 and 2, respectively.

[0227] Table 1. Power grid interaction on typical days

[0228]

[0229] Table 2 Daily operating costs for typical days

[0230]

[0231] As shown in Tables 1 and 2, compared to optimization only on a single day, optimization across multiple time scales reduces grid interaction power by 9.69% and 15.49% on typical days in winter and summer, respectively; and reduces daily operating costs by 4.24% and 3.24%, respectively. These results demonstrate that as the optimization time scale becomes increasingly refined, the system can schedule multiple devices to participate in regulation, effectively mitigating source-load errors and reducing grid interaction fluctuations; simultaneously, the reduction in daily operating costs demonstrates good economic efficiency.

[0232] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A multi-scale optimization operation method for an integrated energy system considering dynamic characteristics, wherein the integrated energy system includes photovoltaic power generation equipment, solar thermal collectors, gas internal combustion engines, waste heat boilers, gas boilers, heat exchangers, electric chillers, absorption chillers, batteries, hot water storage tanks, and cold water storage tanks, characterized in that: It includes four stages: day-ahead optimization, intraday rolling optimization, ramp / descend optimization, and real-time optimization, specifically: (1) Day-ahead optimization: Based on the day-ahead renewable energy output and user load forecast data, and considering the time-of-use electricity price and the equipment operation constraints in the system, with the goal of minimizing daily operating costs and a time scale of 1 hour, a day-ahead scheduling plan is formulated. The cooling, heating and power inequality equation is established through the load margin, and on this basis, an adjustment benchmark is provided for intraday rolling optimization and real-time optimization. (2) Intraday rolling optimization: Renewable energy output and user load data are predicted and updated using 15-minute time scale; the daily scheduling plan within the calculation time domain is modified using 2-hour rolling calculation time domain. The optimization objective is to minimize the energy purchase cost and equipment output adjustment penalty cost within the rolling calculation time domain, and obtain the intraday scheduling plan within the calculation time domain. (3) Climbing / Descending Optimization: Nested with real-time optimization and intraday rolling optimization, optimization calculation is performed with a 5-minute time scale to correct the intraday scheduling plan in the current calculation time domain; within the calculation time domain of the intraday scheduling plan, considering the dynamic characteristics of slow-response equipment, climbing / descending optimization is performed using a genetic algorithm; the slow-response equipment includes gas internal combustion engines, gas boilers and absorption chillers; The use of genetic algorithms for hill climbing / descent optimization specifically involves: Optimization variable: The ramp / deceleration plan of slow-response equipment. The ramp / deceleration plan of slow-response equipment is composed of the ramp / deceleration modes adopted by the equipment in various time periods. Each equipment adjusts its power according to two modes in each time period within the calculation time domain. Mode 1: Start adjusting at the beginning of the time period until the planned output for the next time period is reached; Mode 2: Start adjusting during the time period and reach the planned output for the next time period at the end of the time period. (4) Real-time optimization: Based on real-time forecast data and slow-response equipment ramp-up plan, the output of equipment other than slow-response equipment in the daily scheduling plan is adjusted in real time. The optimization goal is to minimize the system's cold and hot load deficit penalty and equipment power adjustment penalty within the calculation time domain.

2. The multi-scale optimization operation method for a comprehensive energy system considering dynamic characteristics according to claim 1, characterized in that: There is a feedback process among the four stages: daily optimization, intraday rolling optimization, ramp / descent optimization, and real-time optimization.

3. The multi-scale optimization operation method for a comprehensive energy system considering dynamic characteristics according to claim 2, characterized in that: The feedback process is specifically as follows: After intraday optimization, ramp / descent optimization, and real-time optimization, the optimal real-time scheduling plan for each device within this calculation time domain is obtained; the optimal real-time scheduling plan for the first 15 minutes is executed, while the calculation time domain is rolled forward by 15 minutes; the energy storage status of the energy storage device at the 15th minute is used as the initial constraint condition to start the intraday optimization calculation for the next calculation time domain, and the previous optimization process is repeated continuously.

4. The multi-scale optimization operation method for a comprehensive energy system considering dynamic characteristics according to claim 1, characterized in that: In addition to the constraints of each device, the integrated energy system also needs to meet the following constraints: Electrical balance constraints: In the formula: Let t be the theoretical power generation of the photovoltaic panel during time period t; Let t be the power generation capacity of the gas-fired internal combustion engine during time period t; The electrical load of the integrated energy system during time period t; The power consumption of the electric chiller during time period t; and These represent the battery charging power and discharging power during time period t, respectively. and These represent the electricity purchased and sold by the integrated energy system during time period t. Thermal equilibrium constraint: In the formula: The amount of waste heat recovered by the waste heat boiler during time period t; The upper limit of the ramp rate for the absorption chiller during time period t; The solar collector power during time period t; and The heat storage and heat release power of the hot water storage tank during time period t; The heat load of the integrated energy system during time period t; The heat transfer coefficient of the heat exchanger; The heat consumption of the absorption chiller during time period t; Cold balance constraint: In the formula: The refrigeration power of the absorption refrigeration unit during time period t; The refrigeration power of the electric refrigeration unit during time period t; The cooling load of the system during time period t; and These represent the cooling storage and cooling discharge power of the cold water tank during time period t; Energy storage device start-stop state constraints: In the formula: To optimize the computation time domain in advance; Store energy for the i-th type of energy storage device.

5. The multi-scale optimization operation method for a comprehensive energy system considering dynamic characteristics according to claim 1, characterized in that: The equation establishing the cooling-heating-electricity inequality through load margin is expressed as: In the formula: , , These refer to the total power output for power supply, heating, and cooling of the integrated energy system. This is the margin coefficient.

6. The multi-scale optimization operation method for a comprehensive energy system considering dynamic characteristics according to claim 1, characterized in that: The daily operating cost includes equipment operating cost, gas purchase cost, grid connection cost, and battery aging cost, expressed as: In the formula: For equipment operating costs; For the system's gas purchase cost; For grid interaction costs; Costs associated with battery aging; Among them, equipment operating costs Represented as: In the formula: For PV operating costs; For ICE operating costs; For BAT (bat) operating costs; For HE operating costs; For ST operating costs; Operating costs for GB; WB operating costs; For HST operating costs; For EC operating costs; For AC operating costs; For CST operating costs; The heat load of the integrated energy system during time period t; Heat exchanger heat transfer coefficient; gas purchase cost Represented as: In the formula: For natural gas prices; It is the lower heating value of natural gas; The natural gas power consumed by the gas-fired internal combustion engine during time period t; Δt represents the natural gas power consumed by the gas-fired boiler during time period t; Δt is the calculation time step. Grid interaction cost Represented as: In the formula: and These represent the purchase and sale prices of electricity during time period t, respectively. Battery aging cost Represented as: In the formula: Cost of battery charging and discharging conversion; and These are the battery charging and discharging markers at time t, respectively. and These are the battery charging and discharging indicators at time t, respectively. and Each of the following conditions must be met for the battery charge / discharge state transition constraints: 。 7. The multi-scale optimization operation method for a comprehensive energy system considering dynamic characteristics according to claim 1, characterized in that: The objective function for intraday rolling optimization is: In the formula: Adjust the penalty cost for equipment output; the superscript rq indicates the planned output of each piece of equipment for the day, and rn indicates the optimized output of each piece of equipment for the day. It provides power to the absorption chiller; To provide power for electric chillers; To provide power for gas-fired internal combustion engines; Powering the gas-fired boiler; and The heat storage tank stores and releases heat. and The cold water tank stores and releases cooling power. and The charging and discharging power of the battery; and Purchase and sell electrical power to the power grid; Adjust the penalty coefficient for the power of the absorption chiller; Adjust the penalty coefficient for the power of the electric chiller; The penalty coefficient for adjusting the power of a gas-fired internal combustion engine; Adjust the penalty coefficient for the power output of the gas-fired boiler; Adjust the penalty coefficient for the heat storage and release power of the hot water storage tank; Adjust the penalty coefficient for the cold storage capacity of the cold water tank; Adjust the penalty coefficient for battery charging and discharging power; The penalty factor for adjusting the power exchange between the power grids.

8. The multi-scale optimization operation method for a comprehensive energy system considering dynamic characteristics according to claim 1, characterized in that: The system's minimum penalties for cooling and heating load deficits and equipment power adjustment are expressed as follows: In the formula: and These are the penalty coefficients for cold and heat shortages and the penalty coefficients for equipment power adjustments, respectively. and These are respectively a lack of heat and a lack of cold; This is the sum of the squares of the relative power adjustments of each device; Calculation of heat and cold deficits: Calculation of relative power adjustment for equipment: In the formula: the superscript rn represents the daily optimized output of each device, ss represents the real-time optimized output of the device; the subscript max represents the upper limit of the device output. and The charging and discharging power of the battery; and Purchase and sell electrical power to the power grid; It provides power to the electric refrigeration unit.

Images