A sub-model for optimal scheduling of integrated energy systems considering the matching of time characteristics of multiple energy sources.

Abstract

This invention discloses a time-scale adaptive scheduling model that considers time characteristics and scheduling cycle matching. First, the response speed and capacity of common energy storage devices are determined. Then, based on their different response delay times, the capacity of energy storage devices that can function as backups within different time scales is determined. Next, the time scale is dynamically adjusted through the cooperation of a time-scale adaptive sub-master model to determine the scheduling cycle for different time periods. Finally, energy storage devices are used as virtual backups along with generator backups to undertake different wind power consumption tasks within different time cycles, improving the reliability of integrated energy system scheduling and its ability to cope with wind power uncertainties.

Description

Technical Field

[0001] This invention relates to the fields of wind power consumption and integrated energy system optimization scheduling, specifically to an integrated energy system optimization scheduling sub-model that considers the matching of time characteristics of multiple energy sources. Background Technology

[0002] New energy sources such as wind power and solar power are highly uncertain, and their prediction errors are generally compensated for by controllable units within the grid, such as thermal power units. However, against the backdrop of low-carbon transformation, the proportion of new energy sources such as wind power connected to the grid will increase. By 2050, wind power (44%) and solar power (27%) will dominate the supply of renewable energy in China, and non-fossil energy may account for 70% of final electricity consumption. At that time, conventional power regulation alone will be insufficient to cope with the prediction errors caused by the uncertainty of new energy sources. Therefore, it is necessary to explore new ways to integrate new energy sources.

[0003] Energy storage has always been crucial for the integration of new energy sources, but large-capacity energy storage technology is still under research, and existing energy storage capacity is limited, making it difficult to serve as a means of integrating high-proportion wind power in the future. With the development of integrated energy systems, many new energy forms are gradually coming into focus. The storage capacity inherent in these energy systems provides new avenues for the integration of new energy sources. Currently, equipment with energy storage capabilities includes hydrogen storage tanks in hydrogen energy storage systems, liquid storage tanks in carbon capture devices, and pipelines and heat loads in thermal systems. These devices possess a certain degree of bidirectional regulation capability, which can be used to smooth out fluctuations in new energy sources and compensate for prediction errors. However, the time characteristics of various energy sources differ, posing challenges for interconnection. In grid dispatching, these time characteristics are mainly reflected in the response time of dispatch commands. Grid dispatching is divided into day-ahead dispatching, intraday dispatching, and real-time dispatching according to the command issuance cycle. Day-ahead dispatching commands have a longer issuance cycle, allowing sufficient response time for equipment. Intraday and real-time dispatching commands have shorter issuance cycles, requiring matching with the response times of various energy storage devices. Taking the most widely used energy storage technology as an example, supercapacitors, with the fastest response time, can be directly used for real-time 15-minute dispatch, while batteries, with slightly slower response times, can only be used for hourly intraday dispatch. However, for other forms of energy storage that have emerged recently, the time characteristics are more complex. For example, thermal systems have large inertia and large delays, so they are mostly used for day-ahead dispatch. For hydrogen energy storage systems, research on dispatch is still in its early stages, and most applications are intraday dispatch. Research on carbon capture systems currently focuses mainly on how to reduce carbon emissions, and there is no literature on the direct correlation between carbon emissions and wind power consumption, let alone research on dispatch cycles. In addition, there are some pure energy storage methods, such as phase change energy storage and flywheel energy storage. Currently, there is more research on the devices themselves, but less research on their direct application in dispatch, and generally no research on the impact of their response speed. Therefore, there is no research on the coordination of the time characteristics of these energy storage devices with dispatch cycles. Furthermore, whether it is 4-hour, 2-hour intraday dispatch, or 15-minute real-time dispatch, all are artificially fixed dispatch cycles. If the dispatch instruction issuance cycle is too long, the multi-step prediction error of new energy sources will be too large, and the deviation between the predicted and actual values ​​will lead to a significant imbalance between power generation and consumption. If the dispatch instruction issuance cycle is too short, the unit output adjustment will be too frequent, posing a threat to the stable operation of the system. Some scholars have proposed a method that can adaptively adjust the dispatch cycle, but this method only determines the dispatch cycle based on the wind power prediction error and the reserve provided by the thermal power units.

[0004] Currently, there are numerous energy storage devices connected to the grid, which can serve as virtual backups and work alongside thermal power units to mitigate wind power fluctuations and reduce the impact of wind power forecasting errors. However, the time characteristics of various energy storage devices connected to the grid differ, and determining the dispatch cycle requires considering both the forecasting error range of new energy sources and the response speed of the energy storage devices, ensuring that the forecasting error remains within the range that the energy storage devices can synchronously compensate for. Summary of the Invention

[0005] Purpose of the invention: To address the above-mentioned problems, this paper proposes an integrated energy system optimization scheduling sub-model that considers the matching of time characteristics of multiple energy sources. By analyzing the different response speeds and capacities of common energy storage devices, and then determining whether the scheduling cycle needs to be dynamically adjusted based on the available backup energy storage devices and their capacities in different time periods, the scheduling cycle for different time periods is finally determined. The energy storage devices are combined as virtual backups and conventional unit backups to undertake different wind power consumption tasks in different time periods.

[0006] Technical solution: A time-scale adaptive scheduling model for integrated energy systems considering time characteristic matching, with the following steps:

[0007] Step 1: Establish mathematical models of each energy system in the integrated energy system and analyze their different response delay times;

[0008] Step 2: Establish a multi-step prediction sub-model for wind power;

[0009] Step 3: Establish a comprehensive energy system optimization scheduling sub-model that considers the matching of time characteristics of multiple energy sources;

[0010] Step 4: Simultaneously consider the models established in steps 1-3, establish a time-scale adaptive master model of the integrated energy system, and verify the feasibility and superiority of the scheduling model with the system operating cost as the objective.

[0011] Beneficial Effects: The time-scale adaptive scheduling model proposed in this invention, which considers the time characteristics of multi-energy systems and the matching of scheduling cycles, first determines the response speed and capacity of common energy storage devices, obtaining their different response delay times. Then, based on their different response delay times, it determines the energy storage devices that can function within different time scales and their standby capacity. A wind power prediction sub-model and an integrated energy system optimization scheduling sub-model considering the matching of multi-energy time characteristics are established and linked together through a time-scale adaptive master model. By judging whether the integrated energy system is sufficient to compensate for wind power prediction errors, the time scale is dynamically adjusted to determine the scheduling cycle for different time periods, reducing the adverse impact of wind power prediction errors on the integrated energy system. Finally, energy storage devices are combined as virtual backups and unit backups. With the increasing proportion of wind power, they undertake different wind power consumption tasks within different time cycles, improving the stability of the integrated energy system operation, increasing the reliability of system scheduling, and enhancing the system's ability to cope with wind power uncertainties. Attached Figure Description

[0012] Figure 1 This is a schematic diagram illustrating the time characteristics of different energy systems in this invention;

[0013] Figure 2This is a diagram of the heating network topology in this invention;

[0014] Figure 3 This is a diagram showing the backup capacity that each device in the system can provide under the two schemes of this invention;

[0015] Figure 4 The graphs show the results of wind curtailment / load shedding under the two schemes in this invention.

[0016] Figure 5 This is a diagram showing the output of each device in the system under two different schemes in this invention;

[0017] Figure 6 This invention describes the changes in the scheduling cycle when the rated power of the hydrogen storage device is changed.

[0018] Figure 7 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0019] The invention will now be further explained with reference to the accompanying drawings.

[0020] Step 1: Establish mathematical models of each energy system in the integrated energy system and analyze their different response delay times;

[0021] Step 1.1: Solve the mathematical model of the thermal system response, including the time delay characteristics of the heating network and the thermal inertia of the heating area:

[0022]

[0023] In the formula: τ i ρ represents the heat transfer delay time through pipe i, in hours (h); ρ is the density of hot water, in tons (t / m³). 3 L i d is the length of pipe i, in meters; i Here is the inner diameter of pipe i, in meters (m). i Let be the flow rate of pipe i.

[0024]

[0025] In the formula: T t room This represents the indoor temperature of the heating area at time t, in °C. C' represents the total heat dissipation of all radiators within the heat dissipation area at time t, in MW; C' represents the heat capacity per unit heating area, in GJ / m². 2 ·℃; S is the area of ​​the heating area, in m² 2 k1, k2, and k3 are corresponding coefficients. Step 1.2: Solve the mathematical model of the response delay of the hydrogen energy storage system:

[0026] Because hydrogen fuel cells (HFCs) require a heating process to reach their optimal operating state from room temperature, their output voltage and power will also change. Based on the law of conservation of energy, the dynamic heat transfer model of the fuel cell stack is established as follows:

[0027]

[0028]

[0029] In the formula: C t P is the heat capacity of the fuel cell stack; I is the current, which is the product of the current density and the effective cell area; total P elec Q cool Q loss These represent the total power of the fuel cell stack, the output power of the fuel cell stack, the heat removed by the cooling material per unit time, and the heat radiated outward by the fuel cell stack per unit time, respectively; ΔH is the enthalpy of hydrogen; h cond h conv —These represent parameters describing the heat conduction and convection characteristics of a heat exchanger, respectively; T in T out These are the inlet and outlet temperatures of the cooling material, respectively; T, T amb The fuel cell stack temperature and ambient temperature are respectively; R t This represents the thermal resistance of the fuel cell stack.

[0030] The optimal operating temperature of the fuel cell stack at the corresponding load current can be approximated as follows:

[0031]

[0032] In the formula: T target The optimal temperature for the fuel cell stack to operate at the corresponding load current; T min Minimum operating temperature; T max Maximum operating temperature; I max This is the maximum current.

[0033] When a hydrogen fuel cell starts up, the relationship between the stack operating temperature and time can be simplified as follows:

[0034]

[0035] In the formula: D is the delay coefficient of the dynamic response of the hydrogen fuel cell.

[0036] Therefore, when the load current changes abruptly, the output power P of the HFC... FC It can be represented as:

[0037]

[0038] In the formula: N is the number of fuel cell cells, and V is the output voltage of a single cell.

[0039] Step 1.3: Solve the mathematical model of the carbon capture system response:

[0040] The energy time-shifting characteristic of carbon capture power plants involves adjusting the net output of the plant by continuously increasing or decreasing the energy consumption of the carbon capture equipment. The mathematical model for a carbon capture power plant is expressed as follows:

[0041]

[0042] In the formula: P D,t Energy consumption for carbon capture and fixation; P Y,k,t E represents the carbon capture energy consumption of thermal power unit k at time t. ingCO2,k,t λ represents the CO2 processing capacity of thermal power unit k during time period t; λ represents the energy consumption required to process a unit of CO2; e g,k η is the carbon emission intensity of thermal power unit k; β is the carbon capture efficiency; η is the maximum operating condition coefficient of the regeneration tower and compressor; δ k,t E represents the flue gas split ratio of thermal power unit k during time period t; G,k,t E represents the total CO2 emissions of thermal power unit k during time period t; CG,k,t E represents the amount of CO2 supplied to the liquid storage tank of thermal power unit k during time period t. GJ,k,t This represents the net CO2 emissions of thermal power unit k during time period t.

[0043] From the above formula, the relationship between the net output power and the total power of a thermal power unit can be obtained as follows:

[0044] P CONJ,k,t =(1-λδ) k,t βe g,k )P CON,k,t -P D,t -λE CG,k,t

[0045] The faster response speed of carbon capture power plants is related to the energy acquisition method of the carbon capture equipment. The carbon capture equipment can change its operating state by adjusting the steam extraction rate or using plant power, thereby changing the net output of the carbon capture power plant. Compared with the 5-10 minute standby response of conventional thermal power plants, the response of carbon capture power plants can be within 5 minutes.

[0046] Step 1.4: Analyze the response time of each energy system:

[0047] The response time of electricity is on the order of milliseconds, and its response delay is almost negligible. As the above analysis shows, compared to electrical energy, all energy sources have a response time, meaning they all have inertia. The time characteristics of different energy systems vary, such as... Figure 1As shown. Furthermore, the response time of energy is related to the energy type and capacity. For example, in a thermal system, if the heat load is a building, a large capacity means a large building size, which in turn means a large thermal inertia. Additionally, a long primary pipeline means a large thermal delay, all of which contribute to a prolonged response time in the thermal system. Generally speaking, the response time of a thermal system is on the order of hours, and the heat load cannot participate in a 15-minute real-time scheduling. For hydrogen storage systems, the chemical process of electro-hydrogen production is very fast, and its response time is generally not considered. The main concern is the response time of the fuel cell, which is primarily the time it takes for hydrogen to flow through each fuel cell stack. The larger the capacity, the more fuel cells there are, and the longer the delay time, roughly several minutes to tens of minutes. Carbon capture power plants can change their operating state by adjusting the steam extraction rate, thus achieving a faster adjustment rate; the response time of a carbon capture power plant can be within 5 minutes.

[0048] Step 2: Establish a multi-step prediction sub-model for wind power;

[0049] Step 2.1: Use a statistical model to learn the relationships between historical wind power data and construct a sub-model for wind power prediction:

[0050]

[0051] In the formula: t is the current time; Let f be the predicted value at time t+T; f() is the prediction model; x t This refers to measurement data prior to the current time (including the current time).

[0052] Due to the development of deep learning technology, numerous neural networks have been applied to wind power forecasting to mine the temporal relationships of wind power data and address wind power uncertainties. This invention employs a prediction model based on Variational Mode Decomposition-Long Short Term Memory (VMD-LSTM) for wind power forecasting. This model uses the VMD method to decompose the wind power signal into three modal components: long-term, fluctuating, and random. Then, LSTM is used to perform deep learning on each of these three modal components, utilizing its unique forget and remember gate structures to establish correlations between time series over longer time intervals, achieving relatively accurate multi-step predictions.

[0053] Step 3: Establish a comprehensive energy system optimization scheduling sub-model that considers the matching of time characteristics of multiple energy sources;

[0054] Step 3.1: Establish the objective function C for the operating cost of the integrated energy system and solve it separately:

[0055] C = C CON +C CHP +C T +CZ +C QF +C QFH

[0056] In the formula: C represents the total system cost; C CON For the operating cost of thermal power units, C CHP For the operating costs of combined heat and power (CHP) units; C T For carbon trading costs; C Z C is the depreciation cost of carbon capture equipment. QF C. The cost of wind curtailment penalties; QFH The cost is levied to reduce load. Specifically:

[0057]

[0058] Where: N1 is the number of thermal power units; N2 is the number of combined heat and power units; t total For the total scheduling period; a k b k c k α is the operating cost coefficient for thermal power unit k; m β m γ m δ m θ m μ m c1 is the operating cost coefficient of the thermal power unit m; c2 is the price of coal for power generation; c2 is the carbon trading price; ζ QF ζ is the wind curtailment penalty coefficient. QFH P is the load shedding penalty factor; CON,k,t P represents the electrical output of thermal power unit k at time t. CHP,m,t Q is the electrical output of CHP unit m at time t; CHP,m,t P represents the thermal output of the CHP unit at time t; wind,t Let t be the wind power output at time t; and These represent the upper and lower limits of wind power output at time t; E c λ represents the system's total carbon emissions for the entire day. h ω is the carbon quota coefficient for thermal power units; C is the depreciation rate. ZJ N represents the total price of the carbon capture equipment excluding the storage tank. T The depreciation period for carbon capture equipment other than storage tanks; P CY Price per unit volume of liquid storage tank; V CY Storage tank volume; N C This refers to the depreciation period of the liquid storage tank.

[0059] Step 3.2: While solving the objective function C, a series of constraints need to be satisfied:

[0060] The system power balance constraints satisfy both electrical power balance and thermal power balance. The electrical power balance constraints satisfy:

[0061]

[0062] In the formula: P CONJ,k,t P represents the net output power of thermal power unit k during time period t; FC,t P represents the power of the hydrogen fuel cell at time t. EC,t P represents the hydrogen production capacity of the hydrogen production equipment at time t. LD,t P represents the total electrical load power at time t. n,t This refers to the power stored and discharged by the energy storage device.

[0063] The heat power balance state of the heating network is described using the heat network equilibrium equation, which satisfies the following:

[0064]

[0065] In the formula: A1 is the starting point correlation matrix of the heating network topology, and A2 is the ending point correlation matrix of the heating network topology; and These are vectors representing the inflow and outflow heat of each pipe at time t; Q t It represents the net heat flowing into each node at time t, with heat inflow being positive and heat outflow being negative.

[0066] Conventional unit operating constraints mainly include unit output constraints, unit ramp-up constraints, and thermoelectric coupling constraints for CHP units. Among these, the unit output constraints satisfy:

[0067]

[0068] In the formula: These are the lower and upper limits of the electrical power of thermal power units, respectively. These are the lower and upper limits of the electrical power of the CHP unit; These are the lower and upper limits of the thermal power of the CHP unit; These represent the lower and upper limits of wind power output, both in MW.

[0069] The ramping constraints for thermal power units and CHP units are satisfied:

[0070]

[0071] In the formula: These refer to the landslide and ramp power of thermal power units, respectively. These are the landslide and ramp power of the CHP unit, respectively, in MW / h.

[0072] The thermoelectric coupling constraints of the CHP unit are satisfied:

[0073]

[0074] In the formula: C m c is the thermoelectric ratio of the CHP unit under back pressure conditions; m K represents the heat-to-power ratio of the CHP unit m under maximum condensing conditions. m It is a constant.

[0075] In addition to meeting the network delay constraints and the thermal inertia constraints of the heating area, the thermal system should also meet the thermal energy attenuation constraints:

[0076]

[0077] In the formula: The outlet temperature of pipe i after the delay; The inlet temperature of pipe i at time t; T t o ε is the ambient temperature at time t, in °C; ε ​​is the heat conversion coefficient per unit length of the heating network pipe, in GJ / h·m·K; c is the specific heat capacity of the hot water, in GJ / (m³). 2 ·℃); This represents the heat loss of pipe i during the delay period, expressed in MW.

[0078] The heat storage and release potential of the heating network is simulated using a virtual heat storage tank. The simulated heat storage and release meets the following conditions regarding the network temperature:

[0079]

[0080] In the formula: V represents the heat storage and release power of pipeline i after the delay, in MW; Δt represents the duration of the scheduling period, in hours; V i Let i be the volume of pipe i, in meters. 3 .

[0081] The relationship between the heat generated in the pipe and the temperature of the hot water inside the pipe satisfies:

[0082]

[0083] In the formula: Let be the amount of heat flowing into pipe i at time t. Let be the heat flowing out of pipe i at time t.

[0084] The temperature of the hot water in the pipes should also be limited to a certain range to meet the following requirements:

[0085]

[0086] In the formula: Let T be the inlet and outlet temperatures of the water supply pipeline at time t; gin，min T g in，max T g out，min T g out，max These are the upper and lower limits of its inlet and outlet temperatures, respectively; Let T be the inlet and outlet temperatures of the return water pipe at time t; h in，min T h in，max T h out，min T h out，max These are the upper and lower limits of its inlet and outlet temperatures, respectively, with the temperature unit being °C.

[0087] The outflow temperature at the confluence node is calculated using the inflow temperature and inflow flow rate, satisfying the following conditions:

[0088] (∑m out )T out =∑(m in T in )

[0089] Where: m out m in These represent the flow rates of each pipe flowing into and out of the mixing node, in t / h; T in T out These represent the temperatures of each pipe at the inflow and outflow mixing node, respectively, in °C.

[0090] In carbon capture power plants, the energy consumption for carbon capture is constrained by the maximum operating conditions of the regeneration tower and compressor. Therefore, there is a maximum value for the amount of CO2 being processed, and the amount of CO2 that can be extracted from the storage tank satisfies the following:

[0091]

[0092] When considering the capacity of the storage tank, the volume of the stored solution and the mass of CO2 need to be taken into account, and the following conditions must be met:

[0093]

[0094]

[0095] In the formula: V CA,k,t M is the volume of solution required by the storage tank of thermal power unit k to release CO2 during time period t; MEA M is the molar mass of MEA; CO2 θ represents the molar mass of CO2; θ represents the amount of CO2 that the regeneration tower can dissolve (the decrease in CO2 content from the rich solution to the lean solution); μ R σ represents the solution concentration;R V is the solution density; CFL,k,t V represents the volume of solution in the rich liquid tank of thermal power unit k during time period t; CPL,k,t V represents the volume of lean liquor in the boiler's reservoir during time period t; CR,k The capacity of the liquid storage tank configured for thermal power unit k; V CFL,k,start V represents the initial rich liquid tank solution volume of thermal power unit k; CPL,k,start V represents the initial lean solution volume in the lean liquor tank of thermal power unit k; CFL,k,end V represents the volume of solution in the rich liquid tank at the end of the k-th dispatch cycle of the thermal power unit; CPL,k,end This represents the amount of lean solution in the reservoir at the end of the k-th dispatch cycle of the thermal power unit.

[0096] The power of hydrogen production equipment in a hydrogen energy storage system is limited by its rated power.

[0097] 0≤P EC,t ≤P ECN

[0098] In the formula: P EC,t P represents the hydrogen production capacity of the hydrogen production equipment at time t. ECN This refers to the rated power of the hydrogen production equipment.

[0099] The response time of HFC varies under different load rates. When HFC participates in scheduling on a timescale of 15 minutes to 1 hour:

[0100] 0≤P FC,t ≤0.5P FCN

[0101] When HFC participates in the development of scheduling plans on a timescale of 1h-4h:

[0102] 0≤P FC,t ≤P FCN

[0103] In the formula: P FC,t P represents the power of the HFC at time t. FCN This is the rated power of the HFC.

[0104] Hydrogen energy storage systems have internal energy coupling links that satisfy coupling constraints:

[0105]

[0106] In the formula: E H2,in,t E represents the amount of hydrogen stored in the hydrogen storage tank at time t. H2,out,t P represents the amount of hydrogen output from the hydrogen storage tank at time t. H2,in,t and P H2,out,t η represents the input and output power of the hydrogen storage tank at time t. H2,in and η H2,out The input and output efficiency of hydrogen.

[0107] In the energy storage device, the constraints on the storage and discharge command X are satisfied as follows:

[0108]

[0109] In the formula: X represents the energy storage and discharge command of the energy storage device, X=1 represents energy storage, and X=-1 represents discharge.

[0110] The constraints on energy storage and discharge power and energy storage device capacity are satisfied as follows:

[0111]

[0112] In the formula: and The storage / discharge command of the energy storage device at time t; and The storage / discharge power of the energy storage device; and E represents the maximum and minimum storage and discharge power available within a single scheduling period. n,t , and Let η be the energy storage capacity, maximum and minimum available capacity of the energy storage device at time t; chu and η f The charging and discharging efficiency of the energy storage device.

[0113] Step 4: Simultaneously consider the models established in steps 1-3, establish a time-scale adaptive master model of the integrated energy system, and verify the feasibility and superiority of the scheduling model with the system operating cost as the objective.

[0114] Step 4.1: Establish an available backup model for an integrated energy system that considers multiple energy storage backups:

[0115]

[0116] In the formula: Backup provided for the thermal power unit at time t; The backup provided for the CHP unit at time t; Reserved for thermal inertia at time t; Backup provided for the carbon capture equipment at time t; Backup provided for hydrogen storage equipment and HFC at time t; The backup power provided to the energy storage device at time t.

[0117] Step 4.2: Establish the time-scale adaptive master model:

[0118] 1) Select a segment of raw wind power data, set the prediction step size L=1 and the scheduling time scale T=15min as the initial scenario, and perform wind power prediction and comprehensive energy system optimization scheduling considering the time characteristics of each energy storage device.

[0119] 2) Fit historical data prior to the selected original data and specify the confidence level to obtain the wind power statistical prediction error.

[0120] 3) Determine whether the available reserves provided by the system after the dispatch is sufficient to compensate for the wind power statistical prediction error. If the available reserves are insufficient, it indicates that the wind power prediction error is too large at the 15-minute time scale. In this case, other methods to reduce the prediction error in a short time scale need to be considered. The prediction step number L and the time scale T remain unchanged, and the dynamic adjustment ends. If the available reserves of the system are sufficient to compensate for the wind power prediction error, it indicates that the system is sufficient to cope with the adverse effects of the wind power prediction error at this scale. Then, let L+1 and T+15min to reduce the number of unnecessary dispatch instructions issued. Until L (note: in this paper, 1≤L≤16 is set) increases to a certain appropriate value, the wind power prediction error within the dispatch cycle is greater than the available reserves of the system, and the dynamic adjustment ends.

[0121] 4) Repeat the above steps for the next time scale until the last time scale is adjusted, determine the daily scheduling cycle division result, and complete the daily rolling scheduling.

[0122] Step 4.3: Scene Setup:

[0123] The integrated energy system in this invention consists of a power system, a thermal system, and energy storage devices such as carbon capture equipment, a hydrogen energy storage system, and energy storage devices. The power system includes a carbon capture power plant with two thermal power units equipped with carbon capture devices; a pure thermal power plant with two pure condensing thermal power units; a combined heat and power plant with two combined heat and power units; and a wind farm. The thermal system includes two heating zones, supplied by two combined heat and power units. The original wind power data used in this paper comes from publicly available operating data from Elia, the Belgian power operator, with a 90% confidence interval for wind power output. The output range of each unit is shown in Table 1. The calculation results considering different response delay time constants of each energy storage device in the example are shown in Table 2. The heating network adopts a quality regulation method, and its topology is as follows: Figure 2 In the diagram, ① and ② are heat source nodes, ⑦ and ⑧ are heat load nodes, and ③-⑥ are confluence / diversion nodes. Pipes 1, 3, 5, 7, and 9 are water supply pipes, and 2, 4, 6, 8, and 10 are return water pipes. The upper and lower temperature limits for the water supply pipes are set at 120℃ and 100℃, respectively; the upper and lower temperature limits for the return water pipes are set at 80℃ and 60℃, respectively; the indoor design temperature is taken as 20±2℃. The heat conversion coefficient is 1.6×10⁻⁵ GJ / h·m·K.

[0124] Table 1 Output range of each unit

[0125] unit Electric output range / MW Thermal output range / MW CHP1 [100,200] [0,300] CHP2 [100,200] [0,300] CON1 [50,100] [0,0] CON2 [100,200] [0,0] CON3 [100,200] [0,0] CON4 [50,100] [0,0]

[0126] Table 2 Time Characteristics of Some Energy Storage Devices

[0127] Energy storage devices Response latency / h heating network pipelines 0.25 Heating area 0.85 fuel cells 1

[0128] Based on the above parameters, the yalmip+cplex solver is called in the MATLAB environment. The total scheduling time is 1 day and the scheduling interval is 15 minutes. The following two calculation examples are set up:

[0129] Case 1: Fixed time scale (4h) scheduling;

[0130] Case 2: Time-scale adaptive scheduling.

[0131] Simulation results:

[0132] Analysis of the compensation effect of the reserve capacity provided by each part of the integrated energy system on wind power forecasting errors:

[0133] Before and after dynamic adjustment of the time scale, the compensation effect of the reserve capacity provided by each part of the integrated energy system on wind power forecasting errors is as follows: Figure 3 As shown, under a fixed time scale, even with the addition of carbon capture, hydrogen storage, and energy storage devices, the system's available reserves are still insufficient to compensate for wind power prediction errors during the periods of 1:45-3:15, 6:45-7:15, 9:30-11:15, and 19:15-19:45. During these periods, the system's ability to cope with wind power uncertainties is weak. However, after adaptive adjustment of the time scale, the total available reserves of the integrated energy system are sufficient to compensate for wind power prediction errors in any given period, reducing the adverse effects of wind power uncertainties on the system.

[0134] At the same time Figure 3 As shown in (b), during the time-scale adaptive adjustment, the backup provided by conventional turbines alone is insufficient to compensate for wind power prediction errors during certain periods, such as 6:30-7:00, 9:30-11:00, 18:15-19:00, and 21:00-21:30. This also reflects that as the proportion of wind power increases, the uncertainty of wind power increases, and the prediction error of wind power becomes too large. Under this premise, even with time-scale adaptive adjustment, it is still insufficient to cope with the uncertainty of wind power using only conventional turbines; additional energy storage devices are needed for supplementation and assistance. As the proportion of wind power output increases, the available backup in future integrated energy systems will gradually rely mainly on the power support provided by various energy storage devices.

[0135] It is important to note that after a new dispatch instruction is issued, the distance between each dispatch time and the instruction issuance time within the next dispatch cycle shortens. For the wind power forecasting phase, the shortened forecast lead time at the predicted time leads to predicted wind power values ​​that are closer to actual values, thus reducing forecast errors. Therefore, as the time scale changes and the dispatch instruction issuance time changes, the wind power forecast error at some dispatch times will also change accordingly, as reflected in… Figure 3 The curve describing the changes in wind power prediction error exhibits a periodic trend.

[0136] Analysis of wind curtailment / load shedding:

[0137] Figure 4 This illustrates the changes in wind curtailment / load shedding power before and after time-scale adaptive adjustment. As the distance between the scheduling time and the command issuance time increases, the prediction error for wind power often becomes larger in certain periods at the end of the scheduling cycle. Therefore, when the system maintains a constant time scale for scheduling, insufficient system reserve is likely to occur at the end of the scheduling cycle, resulting in significant wind curtailment and load shedding. Figure 4 It can be intuitively seen that after the time scale is adaptively adjusted, the wind curtailment and load shedding power generated by the system scheduling are significantly reduced, and the system's ability to cope with the uncertainty of wind power is greatly improved.

[0138] Analysis of scheduling cycle segmentation:

[0139] As the proportion of wind power output increases, the uncertainty of wind power also increases, leading to greater prediction errors. To ensure that the system's available reserves are sufficient to compensate for wind power prediction errors at any given time, the number of dynamic adjustments to the time scale increases. Table 3 shows a comparison of the system's scheduling cycle and the amount of wind curtailment and load shedding generated in each scheduling cycle before and after adaptive time scale adjustment. As shown in the table, in Case 1, the system rolled over the schedule every 4 hours, issuing a total of 6 scheduling commands to complete the daily rolling scheduling plan. In Case 2, the system adaptively adjusted the time scale by determining whether the system's available reserves were sufficient to compensate for wind power prediction errors, issuing a total of 11 scheduling commands to complete the daily rolling scheduling plan. The amount of wind curtailment and load shedding generated in each scheduling cycle also decreased, improving system stability and scheduling reliability.

[0140] Table 3 Comparison of scheduling cycle and wind curtailment / load shedding power before and after time-scale adaptive adjustment.

[0141]

[0142] Analysis of the output of various equipment in the integrated energy system:

[0143] Figure 5This figure shows the output of each unit in the integrated energy system before and after dynamic adjustments on a time scale. As can be seen from the figure, when wind power accounts for a large proportion of output, thermal power units have lower flexibility and are more limited by the upper and lower limits of unit output. In this case, additional carbon capture equipment is added, and liquid storage tanks are used to undertake some regulation functions. However, the capacity of the liquid storage tanks is limited, making it difficult to handle significant load smoothing. In addition, the hydrogen energy storage system and electric energy storage devices added to the system also undertake some wind power consumption tasks, increasing the flexibility of the integrated energy system. The figure shows that the electric energy storage devices have shorter charging and discharging times and respond more quickly to dynamic adjustments on a time scale, while the hydrogen energy storage system involves internal energy conversion processes, resulting in a relatively slower response speed.

[0144] Analysis of the impact of energy storage system capacity changes on scheduling cycles:

[0145] Taking a hydrogen energy storage system as an example, only changing the rated power of the hydrogen storage equipment to participate in system scheduling yields the following scheduling results: Figure 6 As shown in the figure, since the power of the hydrogen storage device accounts for a relatively small proportion in the example in this paper, the scheduling cycle only shifts slightly in T2, T9, and T10. However, it can be reasonably inferred from this that if the rated power of other energy storage devices is changed at the same time, the scheduling cycle will be more significantly affected by the capacity of the energy storage system.

[0146] The above-described embodiments are preferred embodiments of the present invention and are only used to help illustrate the present invention. The preferred embodiments do not describe all details exhaustively, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations can be made based on the content of this specification. Any obvious improvements, substitutions, or configuration changes that can be made by those skilled in the art without departing from the essence of the present invention are within the protection scope of the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A time-scale adaptive scheduling model considering the matching of time characteristics and scheduling cycle, characterized in that, Includes the following steps: 1) Establish mathematical models of response delay for each energy system, and analyze the response delay time based on the mathematical models; 2) Establish a multi-step prediction sub-model for wind power; 3) Establish a sub-model for the optimal scheduling of the integrated energy system; 4) Simultaneously consider the multiple models established in steps 1)-3), establish a time-scale adaptive scheduling model for the integrated energy system, and verify the feasibility and superiority of the scheduling model with the operating cost of the integrated energy system as the objective. The implementation process of step 1) is as follows: 1.1): Solve the mathematical model of the response delay of the thermal system. The mathematical model of the response delay of the thermal system includes the delay characteristics of the heating network and the thermal inertia of the heating area. The delay characteristics of the heating network are as follows: In the formula: τ i ρ represents the heat transfer delay time through pipe i, in hours (h); ρ is the density of hot water, in tons (t / m³). 3 ; L i d is the length of pipe i, in meters; i Here is the inner diameter of pipe i, in meters (m). i Here is the flow rate of pipe i, in t / h; The thermal inertia of the heating area is: In the formula: k1, k2, k3 are coefficients; T t room This represents the indoor temperature of the heating area at time t, in °C. T represents the total heat dissipation of all radiators within the heat dissipation area at time t, expressed in MW. t o Δt represents the ambient temperature at time t; Δt represents the scheduling interval; C' represents the heat capacity per unit heating area, in GJ / m². 2 ·℃; S is the area of ​​the heating area, in m² 2 μ is the indoor heat loss coefficient; 1.2): Obtain the response delay of the hydrogen energy storage system: Since hydrogen fuel cells (HFCs) require a heating process to reach their optimal operating state from room temperature, their output voltage and power will also change. Based on the law of conservation of energy, the dynamic heat transfer response delay model of the fuel cell stack is established as follows: In the formula: C t I is the heat capacity of the fuel cell stack; P is the current; total P elec Q cool Q loss These represent the total power of the fuel cell stack, the output power of the fuel cell stack, the heat removed by the cooling material per unit time, and the heat radiated outward by the fuel cell stack per unit time, respectively; F is the Faraday constant; ΔH is the enthalpy of hydrogen; V cell The voltage of a single fuel cell; h cond h conv These represent the parameters describing the heat conduction and convection characteristics of a heat exchanger, respectively; T in T out These are the inlet and outlet temperatures of the cooling material, respectively; T, T amb The fuel cell stack temperature and ambient temperature are respectively; R t For the thermal resistance of the fuel cell stack; The response delay of the hydrogen energy storage system is obtained by solving the dynamic heat transfer response delay model of the fuel cell stack. The optimal temperature control method for hydrogen fuel cells (HFCs) is adopted to control power output; the optimal temperature for the fuel cell stack to operate at the corresponding load current is expressed as follows: In the formula: T target The optimal temperature for the fuel cell stack to operate at the corresponding load current; T min Minimum operating temperature; T max Maximum operating temperature; I max This is the maximum current; When a hydrogen fuel cell (HFC) starts up, the relationship between the stack operating temperature and time is simplified as follows: In the formula: D is the delay coefficient of the dynamic response of the hydrogen fuel cell; i is the current density; Therefore, when the load current changes abruptly, the output power P of the hydrogen fuel cell (HFC) will change. FC Represented as: In the formula: N is the number of fuel cell cells, and V is the output voltage of a single cell; 1.3): Obtain the response delay time of the carbon capture power plant: The energy time-shifting characteristic of a carbon capture power plant is achieved by continuously increasing or decreasing the energy consumption of the carbon capture equipment, using the varying energy consumption of the equipment to adjust the net output of the carbon capture power plant. First, the mathematical model of the carbon capture power plant is expressed as follows: In the formula: P CON,k,t Let P be the electrical output of thermal power unit k at time t. CONJ,k,t P represents the net output power of thermal power unit k during time period t; D,t Energy consumption for carbon capture and fixation; P Y,k,t Let be the carbon capture energy consumption of thermal power unit k at time t; λ represents the CO2 processing capacity of thermal power unit k during time period t; λ represents the energy consumption required to process a unit of CO2; e g,k η represents the carbon emission intensity of thermal power unit k; β represents the carbon capture efficiency; and η represents the maximum operating condition coefficient of the regeneration tower and compressor. δ represents the maximum output power of thermal power unit k; k,t E represents the flue gas split ratio of thermal power unit k during time period t; G,k,t E represents the total CO2 emissions of thermal power unit k during time period t; CG,k,t E represents the amount of CO2 supplied to the liquid storage tank of thermal power unit k during time period t. GJ,k,t Let be the net CO2 emissions of thermal power unit k during time period t; From the above formula, the relationship between the net output power and the total power of a thermal power unit can be obtained as follows: P CONJ,k,t =(1-λδ k,t b g,k )P CON,k,t -P D,t -λE CG,k,t Based on research and analysis of carbon capture power plants, the faster adjustment speed of carbon capture power plants is related to the energy acquisition method of carbon capture equipment. Carbon capture equipment can change the operating status of carbon capture equipment by adjusting the steam extraction rate or using plant power, thereby changing the net output of carbon capture power plants. Compared with the 5-10 minute standby response of conventional thermal power plants, the response delay time of carbon capture power plants is within 5 minutes. The implementation process of step 2) is as follows: Constructing a wind power prediction sub-model: In the formula: t is the current time; Let f be the predicted value at time t+T; f() is the variational mode decomposition-long short-term memory neural network prediction model; x t This includes measurement data prior to the current time. The implementation process of step 3) is as follows: 3.1): Establish the objective function C for the operating cost of the integrated energy system: C=C CON +C CHP +C T +C Z +C QF +C QFH In the formula: C represents the total cost of the integrated energy system; C CON For the operating cost of thermal power units, C CHP CHP operating costs for combined heat and power units; C T For carbon trading costs; C Z C is the depreciation cost of carbon capture equipment. QF C. The cost of wind curtailment penalties; QFH To incur cost penalties for load shedding, the specific penalties include: Where: N1 is the number of thermal power units; N2 is the number of CHP units in combined heat and power (CHP); t total For the total scheduling period; a k b k c k α is the operating cost coefficient for thermal power unit k; m β m γ m δ m θ m μ m c1 is the operating cost coefficient of the thermal power unit m; c2 is the price of coal for power generation; c2 is the carbon trading price; ζ QF ζ is the wind curtailment penalty coefficient. QFH P is the load shedding penalty factor; CON,k,t P represents the electrical output of thermal power unit k at time t. CHP,m,t Q is the electrical output of the cogeneration unit at time t (CHPm). CHP,m,t P represents the thermal output of the cogeneration unit at time t (CHPm). wind,t Let t be the wind power output at time t; and These represent the upper and lower limits of wind power output at time t; E c λ represents the system's total carbon emissions for the entire day. h ω is the carbon quota coefficient for thermal power units; C is the depreciation rate. ZJ N represents the total price of the carbon capture equipment excluding the storage tank. T For carbon capture equipment other than storage tanks, the depreciation period is P. CY Price per unit volume of liquid storage tank; V CY Storage tank volume; N C The depreciation period for the storage tank; 3.2): While solving the objective function C of the integrated energy system operating cost, a series of constraints must be satisfied: The power balance constraints of the integrated energy system satisfy both electrical power balance and thermal power balance, wherein the electrical power balance constraint satisfies: In the formula: P CONJ,k,t P represents the net output power of thermal power unit k during time period t; FC,t P represents the power of the hydrogen fuel cell at time t. EC,t P represents the hydrogen production capacity of the hydrogen production equipment at time t. LD,t P represents the total electrical load power at time t. n,t The power stored and discharged by the energy storage device; The thermal power balance constraint is satisfied as follows: In the formula: A1 is the starting point correlation matrix of the heating network topology, and A2 is the ending point correlation matrix of the heating network topology; and These are vectors representing the inflow and outflow heat of each pipe at time t; Q t It represents the net heat flowing into each node at time t, with heat inflow being positive and heat outflow being negative; Conventional unit operating constraints include unit output constraints, unit ramp-up constraints, and thermoelectric coupling constraints for combined heat and power (CHP) units; among which, the unit output constraints satisfy: In the formula: P CON P represents the electrical power of the thermal power unit. CHP Q represents the electrical power of the CHP unit. CHP P represents the thermal power of the CHP unit. wind For wind power, These are the lower and upper limits of the electrical power of thermal power units, respectively. The lower and upper limits of the CHP electrical power of the combined heat and power unit; These are the lower and upper limits of the thermal power of the CHP unit; These are the lower and upper limits of wind power output, both in MW. The CHP ramping constraints for thermal power units and combined heat and power units are satisfied as follows: In the formula: These refer to the landslide and ramp power of thermal power units, respectively. These represent the landslide and ramp-up power of the combined heat and power (CHP) unit, respectively, in MW / h; P CON,t P represents the electrical power of the thermal power unit at time t. CHP,t The electrical power of the CHP unit at time t; The combined heat and power (CHP) unit's thermoelectric coupling constraints are satisfied as follows: In the formula: C m CHPm is the heat-to-power ratio of a combined heat and power (CHP) unit under back pressure conditions; m The heat-to-power ratio (CHPm) of a combined heat and power (CHP) unit under maximum condensing conditions; K m It is a constant; The thermal energy decay constraint is satisfied as follows: In the formula: Refers to the time delay τ i The outlet temperature of pipe i afterwards; The inlet temperature of pipe i at time t; T t o The ambient temperature at time t is expressed in °C; ε ​​is the heat conversion coefficient per unit length of the heating network pipe, expressed in GJ / h·m·K; m i,t Let be the flow rate of pipe i at time t; c is the specific heat capacity of the hot water, in GJ / (m³). 2 ·℃); The heat loss of pipe i during the delay period is expressed in MW. The storage and release of heat and the temperature constraints of the pipeline network are satisfied: In the formula: For pipeline i after a delay τ i Subsequent thermal power storage and release, in MW; Δt is the duration of the scheduling period, in hours; V i Let i be the volume of pipe i, in meters. 3 ; The relationship between the heat generated in the pipe and the temperature of the hot water inside the pipe is subject to the following constraints: In the formula: Let be the amount of heat flowing into pipe i at time t. The heat flowing out of pipe i at time t is expressed in MW. The temperature of the hot water in the pipe meets the constraints: In the formula: Let be the inlet and outlet temperatures of the water supply pipeline at time t; These are the upper and lower limits of its inlet and outlet temperatures, respectively; Let be the inlet and outlet temperatures of the return water pipe at time t; These are the upper and lower limits of its inlet and outlet temperatures, respectively, with the temperature unit being °C. The outflow temperature constraint of the bus node is satisfied as follows: (∑m out )T out =∑(m in T in ) Where: m out m in These represent the outflow and inflow flow rates of each pipe at the mixing node, in t / h; T in T out These are the temperatures of each pipe flowing into and out of the mixing node, in °C. The constraint on the amount of CO2 extracted from the storage tank is satisfied as follows: The capacity constraint of the liquid storage tank is satisfied: In the formula: V CA,k,t M is the volume of solution required by the storage tank of thermal power unit k to release CO2 during time period t; MEA The molar mass of MEA; θ is the molar mass of CO2; θ is the amount that the regeneration tower can resolve; μ R σ represents the solution concentration; R V is the solution density; CFL,k,t V represents the volume of solution in the rich liquid tank of thermal power unit k during time period t; CPL,k,t V represents the volume of lean liquor in the reservoir of the thermal power unit k during time period t; CR,k The capacity of the liquid storage tank configured for thermal power unit k; V CFL,k,start V represents the initial rich liquid tank solution volume of thermal power unit k; CPL,k,start V represents the initial lean solution volume in the lean liquor tank of thermal power unit k; CFL,k,end V represents the volume of solution in the rich liquid tank at the end of the k-th dispatch cycle of the thermal power unit; CPL,k,end The amount of lean solution in the liquid tank at the end of the k-dispatch cycle of the thermal power unit; The power constraints of hydrogen production equipment in a hydrogen energy storage system meet the following requirements: 0≤P EC,t ≤P ECN In the formula: P EC,t P represents the hydrogen production capacity of the hydrogen production equipment at time t. ECN The rated power of the hydrogen production equipment; Since the response time of hydrogen fuel cells (HFCs) varies under different load rates, the constraints must be satisfied when HFCs participate in the development of scheduling plans with a timescale of 15 min to 1 h: 0≤P FC,t ≤0.5P FCN When hydrogen fuel cells (HFCs) participate in the scheduling of events on a timescale of 1h-4h, the following constraints must be met: 0≤P FC,t ≤P FCN In the formula: P FC,t P represents the power of the hydrogen fuel cell at time t. FCN This refers to the rated power of the hydrogen fuel cell; Hydrogen energy storage systems have internal energy coupling links, including the coupling between the electrolyzer and the hydrogen storage tank, and the coupling between the hydrogen storage tank and the hydrogen fuel cell. The constraints of these two coupling links satisfy the following: In the formula: E H2,in,t E represents the amount of hydrogen stored in the hydrogen storage tank at time t. H2,out,t P represents the amount of hydrogen output from the hydrogen storage tank at time t. H2,in,t and P H2,out,t η represents the input and output power of the hydrogen storage tank at time t; H2,in and η H2,out The input and output efficiency of hydrogen; The constraints of the storage / discharge command X are satisfied: In the formula: X represents the energy storage and discharge command of the energy storage device, X = 1 represents energy storage, and X = -1 represents discharge; The constraints on energy storage and discharge power and energy storage device capacity are satisfied as follows: In the formula: and The storage and discharge commands of the energy storage device at time t; and For the storage and discharge power of energy storage devices; and E represents the maximum and minimum storage and discharge power available within a single scheduling period. n,t , and Let be the energy storage capacity, maximum and minimum available capacity of the energy storage device at time t; η chu and η f The charging and discharging efficiency of the energy storage device; the implementation process of step 4 is as follows: Step 4.1: Establish an available backup model for an integrated energy system that considers multiple energy storage backups: In the formula: Backup provided for the thermal power unit at time t; The backup provided for the CHP unit at time t; Reserved for thermal inertia at time t; Backup provided for the carbon capture equipment at time t; Backup provided for hydrogen storage equipment and hydrogen fuel cells at time t; Backup provided to the energy storage device at time t; Step 4.2: Establish a time-scale adaptive scheduling model: 1) Select a segment of raw wind power data, set the prediction step size L=1 and the scheduling time scale T=15min as the initial scenario, and perform wind power prediction and comprehensive energy system optimization scheduling considering the time characteristics of each energy storage device; 2) By fitting historical data prior to the original data and specifying the confidence level, the statistical prediction error of wind power is obtained; 3) Determine whether the available reserves provided by the system after the dispatch is sufficient to compensate for the wind power statistical forecast error. If the available reserves of the system are insufficient, it means that the wind power forecast error is too large at the 15-minute time scale. At this time, it is necessary to consider other methods to reduce the forecast error in a short time scale. The number of forecast steps L and the time scale T remain unchanged, and the dynamic adjustment ends. If the available reserves of the integrated energy system are sufficient to compensate for the wind power forecast error, it means that the integrated energy system is sufficient to cope with the adverse effects of the wind power forecast error at this scale. Then, let L+1 and T+15min to reduce the number of unnecessary dispatch instructions. Until L increases to the point that the wind power forecast error in the dispatch cycle is greater than the available reserves of the system, the dynamic adjustment ends. 4) Repeat the above steps for the next time scale until the last time scale is adjusted, determine the daily scheduling cycle division result, and complete the daily rolling scheduling.

Images