Multi-time scale hierarchical scheduling method for combined heat and power system with electric vehicles

Abstract

The application discloses a multi-time scale hierarchical scheduling method for a combined heat and power system containing electric vehicles, and the system comprises a micro gas turbine unit, an absorption refrigeration unit, a ground source heat pump unit, a wind turbine unit, photovoltaic, a battery, a super capacitor and a heat storage device. The optimization method contains two time scales. In the day-ahead stage, economic scheduling is carried out according to the day-ahead predicted power value of wind power and photovoltaic and the day-ahead predicted value of load power in the system, and a target function and optimization constraint condition are established, and the result is used as a reference for the day-ahead optimization. In the day-ahead stage, a two-layer rolling optimization model is considered, a target function and a constraint are established, and a model predictive control (MPC) rolling optimization layer corrects the result of the upper layer according to real-time predicted power information and non-planned behavior of the electric vehicle (EV). The application adopts a multi-time scale hierarchical optimization strategy, real-time tracks the optimal scheduling strategy of the combined heat and power system containing electric vehicles, and realizes a win-win situation between the electric vehicle owner and the micro-grid operator.

Description

Technical Field

[0001] This invention relates to the field of combined heat and power systems, and in particular to a multi-time-scale hierarchical scheduling optimization method that takes into account the conditions for the introduction of electric vehicles into the system, applicable to energy management in integrated energy parks. Background Technology

[0002] Combined heat and power (CHP) systems, as important energy utilization systems in large and medium-sized cities in my country, have solved the problem of centralized urban heating while improving the overall energy efficiency of society. The deployment of CHP systems in cities has received widespread attention worldwide. Considering the randomness and volatility of wind and solar power output and predicted load power values ​​in CHP microgrids, and the fact that their prediction accuracy improves as the time scale decreases, existing microgrid energy management often adopts a multi-time-scale energy management approach that combines day-ahead planning and real-time dispatch.

[0003] In recent years, the sales of electric vehicles (EVs) have accounted for an increasing proportion of the automotive market. Consequently, the issue of integrating EVs into the power grid has become increasingly serious. The integration of EVs into the grid may cause an increase in peak grid load, which is detrimental to the safe and stable operation of the power system.

[0004] In current research on the dispatching of combined heat and power microgrids, the introduction of electric vehicle charging and discharging management systems will bring additional variables to the economic operation of the system and pose a threat to its safe operation.

[0005] The existing technology has the following problems:

[0006] In combined heat and power (CHP) microgrids, wind and solar power output and predicted load power values ​​exhibit randomness and volatility, with prediction accuracy increasing as the time scale decreases. Current microgrid energy management often employs a multi-time-scale approach, combining day-ahead planning and real-time dispatch. Day-ahead planning base values ​​for unit combination and operation are established based on predicted data, while real-time data is used to correct deviations inherited from higher-level systems. Energy management involving electric vehicles (EVs) is often approached from the microgrid operator's perspective, aiming to improve the microgrid's economic efficiency through charging and discharging optimization. However, it's important to note that microgrid operators and EV owners have different interests; therefore, EV owner satisfaction should be considered when developing EV charging and discharging plans. Furthermore, in CHP microgrid energy management, EV emergency charging and unplanned temporary travel increase real-time dispatch pressure, a situation that poses a significant threat to operational economy and safety and should be taken into account. Summary of the Invention

[0007] To overcome the problems existing in the prior art, the present invention aims to provide a multi-time-scale hierarchical scheduling method for a combined heat and power system containing electric vehicles. This method considers both the benefits to microgrid operators and the satisfaction of EV owners when managing the charging and discharging of electric vehicles. Furthermore, by effectively identifying unplanned EV behaviors, the equivalent load of the electric vehicle network is corrected, thereby reducing real-time scheduling pressure and improving the economy and safety of the combined heat and power system.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] Step 1: Establish a combined heat and power microgrid system;

[0010] Step 2: Establish mathematical models of equipment in a combined heat and power microgrid system;

[0011] Step 3: Determine the objective function and day-ahead economic scheduling constraints;

[0012] Step 4: Establish a multi-timescale optimization model for the cogeneration microgrid: The multi-timescale optimization model for the cogeneration microgrid includes two time scales: Day-ahead: economic optimization scheduling of the cogeneration microgrid; Intra-day: joint optimization of the EV charging / discharging management layer and the Model Predictive Control (MPC) rolling optimization layer; The EV charging / discharging management layer takes into account the battery degradation costs caused by the charging and discharging of electric vehicles and establishes a multi-objective optimization scheduling model that minimizes the load curve variance and minimizes the dissatisfaction of EV owners participating in the vehicle-to-grid (V2G) mechanism; When running in the Model Predictive Control (MPC) rolling optimization layer, if the number of unplanned EVs exceeds the threshold, it will return to the EV charging / discharging management layer to solve again and then return.

[0013] As a specific implementation method, in step 1, the equipment in the cogeneration microgrid system includes micro gas turbine units, absorption chiller units, ground source heat pump units, wind turbine units, photovoltaics, batteries, supercapacitors, and thermal storage devices.

[0014] As a specific implementation method, in step 3, the method for determining the objective function and the day-ahead economic scheduling constraints is as follows:

[0015] The day-ahead economic dispatch optimization cost function for the S3.1 microgrid operation is as follows:

[0016] minF day-ahead =F utility +F om +F mt +F e

[0017] In the formula, F utility F represents the power cost of switching with the main network. omIndicates DG maintenance cost, F mt F represents the operating cost of a gas turbine. e This represents the cost of treating pollutant gases, minF day-ahead This represents minimizing the day-ahead economic dispatch cost;

[0018] S3.2 Determine the day-ahead economic dispatch constraints, including power balance constraints, distributed power output constraints, micro gas turbine ramping constraints, microgrid and main grid power exchange constraints, main grid power purchase and sale status mutual exclusion constraints, micro gas turbine start-up and shutdown time constraints, battery operation constraints, and thermal storage device constraints.

[0019] S3.3 Determine the objective function for intraday two-level rolling optimization scheduling:

[0020] Electric vehicle charging and discharging management layer:

[0021] Objective function 1:

[0022]

[0023] In the formula, P load (t) represents the load power value at time t, Mean(P) load () represents the average load power at all times, T is the scheduling period, and minF1 EV This represents minimizing the variance of the total load curve;

[0024] Objective function 2:

[0025]

[0026] In the formula, Let be the degradation cost of the nth electric vehicle at time t. N represents the minimum and maximum degradation costs accepted by the user when signing the agreement, respectively. cv Let T be the total number of electric vehicles and T be the scheduling period. This indicates minimizing user dissatisfaction with participating in connected car initiatives;

[0027] Model Predictive Control (MPC) Rolling Optimization Layer:

[0028] The first phase considers the highest economic efficiency when the combined heat and power microgrid is connected to the grid, and establishes an economic optimization model:

[0029]

[0030] These represent the costs of power exchange with the main grid, distributed power source maintenance, micro gas turbine operation, and pollutant gas treatment in the first phase, respectively. and These are the reference values ​​for the power exchanged with the main grid and the power of the micro gas turbine at time t, obtained from the optimized scheduling a few days ago. and These represent the planned values ​​for the power exchanged with the main grid and the power of the micro gas turbine during the first stage at time t, respectively. and These are the penalty coefficients for adjusting the power exchanged with the main grid and the power of the micro gas turbine in the first stage, respectively; minF1 represents minimizing the total operating cost of the first stage.

[0031] The optimization objective of the second stage is to balance the impact of power prediction errors based on the output of each unit obtained in the first stage.

[0032]

[0033] in F i snd , These represent the power regulation costs of the interconnection line with the main grid, the power regulation costs of the i-th distributed power source, and the operation and maintenance costs of the supercapacitor, respectively. snd Let F2 be the total number of distributed power sources, and minF2 be the minimum second-stage regulation cost.

[0034] S3.4 Intraday two-layer rolling optimization constraints: The electric vehicle charging and discharging management layer includes the upper and lower limits of the electric vehicle charging and discharging power, the charging and discharging state variables of the electric vehicle at a certain moment, and the upper and lower limits of the electric vehicle battery capacity; The Model Predictive Control (MPC) rolling optimization layer satisfies the constraints in S3.2.

[0035] As a specific implementation method, in step S3.2, the day-ahead economic scheduling constraints specifically include:

[0036] Power balance constraints:

[0037] P Batt (t)+P MT (t)+P PV (t)+P WT (t)+P Grid (t)=P Load (t)

[0038] Q AM (t)+Q HP (t)+Q CS (t)=Q load (t)

[0039] In the formula, P Batt (t), P MT (t), P PV (t), PWT (t), P Grid (t), P Load (t) represents the battery power, micro gas turbine power, photovoltaic module power, wind turbine power, power exchanged with the main grid, and electrical load power at time t, respectively. AM (t), Q HP (t), Q CS (t), Q load (t) represents the output thermal power of the absorption chiller, the output thermal power of the ground source heat pump unit, the output thermal power of the thermal storage device, and the thermal load power at time t, respectively.

[0040] Distributed power supply output power constraints

[0041] P DG.i.min ≤P DG.i ≤P DG.i.max

[0042] In the formula, P DG.i P DG.i.min P DG.i.max These are the output power of the i-th distributed power source and the lower and upper limits of its output power, respectively.

[0043] Micro gas turbine ramp constraint

[0044]

[0045] In the formula, R up R down P represents the maximum rate of increase and rate of decrease of the gas turbine's output power. MT (t), P MT (t-1) represents the output power of the micro gas turbine at time t and time t-1, respectively, and Δt is the time step;

[0046] Microgrids exchange power constraints with the main grid

[0047] |P Grid (t)|≤P Grid.max

[0048] In the formula, |P Grid (t)| represents the absolute value of the power exchanged with the main network at time t, P Grid.max This is the upper limit of the power exchanged with the main network;

[0049] Mutual Exclusion Constraints of Main Grid Power Purchase and Sale Status

[0050] U Buy (t)+U Sell (t)≤1

[0051] In the formula, U Buy(t), U Sell (t) represents the status quantity of the main grid's electricity purchase and sale during time period t, with a value of 0 or 1;

[0052] Micro gas turbine start-stop time constraints

[0053]

[0054] In the formula, U t-k+1 For the start-up and shutdown state of the gas turbine during the time period t-k+1, T mup T mdown These are the minimum start-up time and minimum shutdown time of the gas turbine, respectively.

[0055] Battery operating constraints

[0056] SOC min ≤SOC(t)≤SOC max

[0057] -P ch.max ≤P Batt (t)≤P dis.max

[0058] |SOC(N t -SOC(1)|≤ε

[0059] U Batt.ch (t)+U Batt.dis (t)≤1

[0060] In the formula, SOC min and SOC max These correspond to the minimum and maximum allowable SOC values ​​of the battery during operation, respectively; P ch.max P dis.max ε represents the maximum allowable charging and discharging power of the battery; ε is the maximum range of battery SOC change after one scheduling cycle; U Batt.ch (t), U Batt.dis (t) represents the state of charge / discharge of the battery during time period t, with a value of 0 or 1; N t The scheduling period;

[0061] thermal storage device constraints

[0062] λ min S ES ≤S ES (t)≤λ max S ES

[0063] -Q ch.max <Q CS (t)<Q dis.max

[0064] In the formula: SES λ is the rated capacity of the thermal storage device; min , λ max These are the minimum and maximum capacity coefficients of the thermal storage device, respectively; Q ch.max Q dis.max These represent the maximum charging and discharging power of the thermal storage device.

[0065] As a specific implementation method, in step S3.4, the constraints on the upper and lower limits of the electric vehicle's charging and discharging power, the constraints on the charging and discharging state variables of the electric vehicle at a certain moment, and the constraints on the upper and lower limits of the electric vehicle's battery capacity specifically include:

[0066] SOC min ≤SOC n (t)≤SOC max

[0067]

[0068]

[0069] In the formula, This represents the charging and discharging power of the nth electric vehicle during time period t. These represent the charging and discharging statuses of the nth electric vehicle during time period t. SOC n (t) represents the SOC value of the nth electric vehicle at time t. min SOC max These are the lower and upper limits of the State of Charge (SOC) for electric vehicles, respectively.

[0070] As a specific implementation method, the specific process of step S4 is as follows:

[0071] S4.1 In the day-ahead phase: Economic dispatch is carried out based on the obtained day-ahead forecasts of wind power, photovoltaic power output and load power, and the results are used as reference values ​​for intraday rolling optimization;

[0072] The S4.2 EV charging and discharging management layer optimizes the charging and discharging of EVs under various V2G modes based on the historical habit data of vehicle owners, obtains the load data of superimposed electric vehicles, and inputs it into the Model Predictive Control (MPC) rolling optimization layer.

[0073] The first stage of the S4.3 MPC rolling optimization layer is based on T u For the time window, Δt u Economic scheduling is performed for the time step, and the result is passed as a reference value to the second stage S4.4.

[0074] The second stage of the S4.4 MPC rolling optimization layer is based on T l For the time window, Δt lThe V2G load information is updated based on the received real-time forecast data of wind and solar power and the identification of EVs for emergency charging and unplanned travel, thereby coordinating the output of each distributed power source and enabling the power connected to the main grid to track the planned value of the upper level.

[0075] S4.5 When the second stage forward rolls over and Δt is executed u After a certain period of time, return to the first stage, scroll the time window forward, and repeat the above process.

[0076] The intraday phase is divided into two layers: the first layer is S4.2, the EV charge and discharge management layer; the second layer is the MPC rolling optimization layer composed of S4.3 and S4.4.

[0077] The MPC rolling optimization layer is divided into two stages, S4.3 and S4.4. The results of S4.3 provide a reference for the control variables in S4.4, while S4.4 feeds back the actual values ​​of the state variables to S4.3, thus playing a role in feedback correction.

[0078] As a specific implementation method, in step S4, the objective function is solved using a combination of the solver CPLEX and a heuristic algorithm.

[0079] The beneficial effects of this invention are:

[0080] 1. A degradation cost model for electric vehicles was developed, and a multi-objective optimization scheduling model for EV charging and discharging was constructed with the goals of reducing load curve deviation and minimizing EV owner dissatisfaction with V2G participation. This model comprehensively considers the interests of EV owners and microgrid operators, achieving a win-win situation.

[0081] 2. To mitigate the impact of EV owners' temporary changes to their plans (urgent charging needs and unplanned travel) on optimized scheduling schemes, and to further reduce the impact of predicted power errors in the CHP microgrid, this invention proposes a two-layer intraday rolling optimization model based on day-ahead scheduling results. When operating at the MPC rolling optimization layer, in addition to using updated prediction data, unplanned EV behavior can be detected, and the V2G load output by the EV charging and discharging management layer can be corrected, improving the economy and accuracy of scheduling, and enhancing the tracking capability with planned values ​​on the main grid tie line. Attached Figure Description

[0082] Figure 1 This is a schematic diagram of the structure of an embodiment of the cogeneration microgrid of the present invention.

[0083] Figure 2 This is a schematic diagram of the intraday two-layer rolling optimization framework of the present invention.

[0084] Figure 3 This is a flowchart of the multi-time-scale hierarchical optimization scheduling method of the present invention.

[0085] In the diagram: 1-Power grid, 2-Battery and supercapacitor, 3-Wind turbine generator set, 4-Photovoltaic power generation module, 5-User electric heating load, 6-Ground source heat pump unit, 7-Gas storage unit, 8-Micro gas turbine unit, 9-Absorption chiller unit, 10-Heat storage device. Detailed Implementation

[0086] Specific embodiments of the present invention will now be described with reference to the accompanying drawings. It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. The technical features or combinations of technical features described in the following embodiments should not be considered isolated; they are combined to achieve better technical effects.

[0087] like Figure 1 As shown, this embodiment of the invention is a combined heat and power microgrid system, including a micro gas turbine unit 8, an absorption chiller unit 9, a ground source heat pump unit 6, a wind turbine generator unit 3, a photovoltaic power generation module 4, a battery and supercapacitor 2, a thermal storage device 10, a power grid 1, a user electric and thermal load 5, and a gas storage unit 7.

[0088] like Figure 3 As shown in the flowchart of the multi-time-scale hierarchical optimization scheduling method of this invention, it can be generally divided into four steps:

[0089] S1. Establish a combined heat and power microgrid system;

[0090] S2. Establish mathematical models of equipment in a combined heat and power microgrid system;

[0091] S3. Determine the objective function and day-ahead economic scheduling constraints;

[0092] S4. Establish a multi-time-scale optimization model for combined heat and power microgrids;

[0093] As a specific embodiment, in step S1, the main equipment in the combined heat and power microgrid system includes a micro gas turbine unit 8, an absorption chiller unit 9, a ground source heat pump unit 6, a wind turbine generator unit 3, photovoltaic power generation modules 4, a battery and supercapacitor 2, and a thermal storage device 10. This system is connected to the main power grid and can purchase electricity from grid 1 when electricity prices are low and sell electricity to grid 1 when electricity prices are high.

[0094] As a specific embodiment, the main operating mathematical model of each device is established as follows:

[0095] 1. Output Model of Absorption Refrigeration Unit

[0096] Absorption chillers typically use lithium bromide absorption chillers to cool the waste heat discharged from micro gas turbines. Their model is represented as:

[0097]

[0098]

[0099] Q AM (t)=η rec C AM Q MT (t)

[0100] In the formula, P MT The magnitude of the electrical output power of the micro gas turbine; η MT The power generation efficiency of a micro gas turbine; Q MT (t), Q AM (t) represents the waste heat emission of the micro gas turbine and the cooling power of the absorption chiller unit at time t, respectively; η L η is the heat loss coefficient of the micro gas turbine; rec The waste heat recovery rate of the absorption chiller unit; C AM The coefficient of performance (COP) of an absorption chiller unit.

[0101] 2. Output Model of Ground Source Heat Pump Unit

[0102] Ground source heat pumps are a highly efficient and energy-saving air conditioning technology that utilizes shallow geothermal resources to provide both heating and cooling. They are over 40% more energy-efficient than conventional central air conditioning systems and have enormous potential for energy conservation in large public buildings. The model is as follows:

[0103] Q HP (t)=C OP ·P HP (t)

[0104] In the formula: Q HP (t), P HP (t) represents the heat power generated by the ground source heat pump and the electrical power consumed by the ground source heat pump at time t, respectively; C OP This is the coefficient of performance (COP) of a ground source heat pump.

[0105] 3. Energy storage device model

[0106] Energy storage devices include electrical energy storage and thermal energy storage. Thermal energy storage devices use water-cooled storage equipment and have similar operating characteristics to electrical energy storage devices. Their model can be represented as follows:

[0107]

[0108] In the formula: E B (t), E B (t-1) represent the capacity of the energy storage device at times t and t-1, respectively; τ is the energy storage loss coefficient; P B.ch (t), P B.dis(t) represents the charging and discharging power of the energy storage device at time t; η ch η dis These represent the charge / discharge efficiency, respectively.

[0109] 4. Wind turbine output model

[0110] The output power of a wind turbine can be expressed by the following formula:

[0111]

[0112] In the formula P WT Where ρ is the output power of the wind turbine, V is the air density, R is the wind speed, and C is the blade radius of the wind turbine. P This is the wind energy utilization coefficient.

[0113] The relationship between the power output of a wind turbine and wind speed is as follows:

[0114]

[0115] In the formula, P rated For rated power, V rated For the rated wind speed, V cutin To cut off the wind speed, V cutout To cut off the wind speed.

[0116] 5. Photovoltaic power generation module output model

[0117] The output power of a photovoltaic panel can be expressed as follows:

[0118]

[0119] In the formula, P PV P is the output power of the photovoltaic panel. STC S represents the output power of a photovoltaic panel under standard conditions. PV S represents the actual solar radiation intensity. STC The solar radiation intensity under standard conditions is given by k, where k is the temperature coefficient and T is the solar radiation intensity under standard conditions. PV and T ref These are the actual ambient temperature and the ambient temperature under standard conditions, respectively. r η is the area of ​​the photovoltaic panel. PV The efficiency of the photovoltaic panel.

[0120] As a specific embodiment, in step S3, the day-ahead economic optimization objective function is determined:

[0121] Considering that the combined heat and power (CHP) microgrid achieves the highest economic efficiency when connected to the grid, an economic optimization model for microgrid operation is established. This model mainly includes: the cost of power exchange with the main grid, the maintenance costs of photovoltaic, wind turbine, battery, and thermal storage devices, the operating cost of the gas turbine, and the cost of treating pollutants. The objective function is:

[0122] minF day-ahead =F utility +F om +F mt +F e

[0123] The formula includes:

[0124] (1) Cost of power exchange with the main network

[0125]

[0126] In the formula C buy (t), C sell (t) represents the purchase and sale prices of electricity at time t, P Grid_buy (t), P Grid_sell (t) represents the power purchased and sold from the main grid at time t, Δt is the dispatch time step, and N t The scheduling period is [number].

[0127] (2) Distributed power generation (DG) maintenance costs

[0128]

[0129] In the formula, |P i (t)| represents the absolute value of the output power of the i-th distributed power source at time t, K om.i Let be the operation and maintenance coefficient of the i-th distributed power source. Ni This represents the total number of distributed power sources.

[0130] (3) Gas turbine operating costs

[0131]

[0132] C mt f is the fuel cost coefficient for gas turbines, expressed in yuan / m³. mt C represents the gas turbine fuel consumption coefficient, expressed in m³ / kWh, where S(t) represents the start-up and shutdown state of the gas turbine during time period t; mts The starting cost of the gas turbine is expressed in yuan per cycle.

[0133] (4) Cost of treating polluting gases

[0134]

[0135] CeSO2, CeCO2, CeNO x Cost per unit of pollutant gas treatment; Emission SO2(t), Emission CO2(t), Emission NO x (t) represents the amount of pollutant gas emitted during time period t.

[0136] As a specific embodiment, the process of determining the day-ahead economic scheduling constraints in step S3 is as follows:

[0137] The specific constraints on economic regulation currently include:

[0138] 1. Power balance constraint:

[0139] P Batt (t)+P MT (t)+P PV (t)+P WT (t)+P Grid (t)=P Load (t)

[0140] Q AM (t)+Q HP (t)+Q CS (t)=Q load (t)

[0141] 2. Output power constraints of distributed power sources

[0142] P DG.i.min ≤P DG.i ≤P DG.i.max

[0143] The output power of photovoltaic, wind turbines and gas turbines should not exceed their maximum capacity.

[0144] 3. Micro gas turbine ramping constraints

[0145]

[0146] 4. Power exchange constraints between microgrids and the main grid

[0147] |P Grid (t)|≤P Grid.max

[0148] 5. Mutual Exclusion Constraints on Main Grid Power Purchase and Sale Status

[0149] U Buy (t)+U Sell (t)≤1

[0150] U Buy (t), U Sell (t) is the state quantity of the main grid electricity purchase and sale during time period t, and its value is 0 or 1.

[0151] 6. Start-up and shutdown time constraints of micro gas turbines

[0152]

[0153] U t-k+1 For the start-up and shutdown state of the gas turbine during the time period t-k+1, T mup T mdown These are the minimum start-up time and minimum shutdown time of the gas turbine, respectively.

[0154] 7. Battery operating constraints

[0155] SOC min ≤SOC(t)≤SOC max

[0156] -P ch.max ≤P Batt (t)≤P dis.max

[0157] |SOC(N t )-SOC(1)|≤ε

[0158] U Batt.ch (t)+U Batt.dis (t)≤1

[0159] SOC min and SOC max These correspond to the permissible SOC range of the battery during operation; P ch.max P dis.max U represents the maximum allowable charging and discharging power of the battery; ε represents the maximum range of battery SOC change after one scheduling cycle; Batt.ch (t), U Batt.dis (t) represents the state of charge / discharge of the battery during time period t, with a value of 0 or 1.

[0160] 8. Constraints of thermal storage devices

[0161] λ min S ES ≤S ES (t)≤λ max S ES

[0162] -Q ch.max <Q ES (t)<Q dis.max

[0163] In the formula: S ES λ is the rated capacity of the thermal storage device; min , λ max These are the minimum and maximum capacity coefficients of the thermal storage device, respectively; Qch.max Q dis.max These represent the maximum charging and discharging power of the thermal storage device.

[0164] As a specific embodiment, in step S3, the intraday two-layer rolling optimization scheduling objective function is determined:

[0165] 1. Electric vehicle charging and discharging management layer:

[0166] Based on the relationship between the depth of discharge and the number of cycles of a Ni-Cd battery for electric vehicles, the optimal fitting curve formula for the battery life curve is as follows:

[0167] L B (DOD) = a × DOD -b ×e -c|DOD

[0168] In the formula, DOD represents the depth of discharge, and a, b, and c are fitting coefficients. Taking a single discharge event as an example, the average output power of the battery within a time interval Δt is P. B If (t), then the depth of discharge (DOD) during this time interval can be expressed by the following formula:

[0169]

[0170] In the formula E BA (t) represents the actual full capacity of the battery at time t. Based on the definition of depth of discharge and the relationship between depth of discharge and number of cycles, the degradation cost of the battery can be obtained as follows:

[0171]

[0172] In the formula, C re η represents the replacement cost of the battery. Bc η Bd These represent the charging and discharging efficiency coefficients of the battery, respectively. After a charging / discharging event, the actual full capacity of the battery will decrease. The actual full capacity of the battery at time t+Δt can be determined by combining the battery's factory rated capacity E. B.rated Calculate according to the following formula:

[0173]

[0174] Since the degradation cost of a battery within each time interval Δt can only be determined after a charging or discharging event has ended, it is necessary to define the direction of the battery power flow to determine whether a complete charge / discharge event has ended. A binary variable g(t) is used to indicate whether the direction of the battery power flow has changed. The definition of g(t) is as follows:

[0175]

[0176] Based on this variable, the cumulative power is defined as follows:

[0177] E a (t)=(1-g(t))E a (t-1)+P B (t)Δt

[0178] Therefore, in summary, the operating cost expression considering the battery degradation effect is as follows:

[0179]

[0180] Based on this, an objective function for the electric vehicle charging and discharging management layer is proposed.

[0181] Objective function 1:

[0182]

[0183] Objective function 2:

[0184]

[0185] 2. MPC Scrolling Optimization Layer:

[0186] The first phase considers the highest economic efficiency when the combined heat and power microgrid is connected to the grid, and establishes an economic optimization model:

[0187]

[0188] The optimization objective of the second stage is to balance the impact of power prediction errors based on the output of each unit obtained in the first stage.

[0189]

[0190] in F i snd , These represent the power regulation costs of the main grid interconnection line, the power regulation costs of the i-th distributed power source, and the operation and maintenance costs of the supercapacitor, respectively. sn d represents the total number of distributed power sources.

[0191] 1) Power regulation cost of the interconnection line with the main grid:

[0192]

[0193] In the formula This represents the power exchanged with the main network at time t, obtained during the first phase. This represents the planned power exchange value with the main grid at time t in the second phase. This is the penalty coefficient for adjusting the power exchange with the main grid in the second phase.

[0194] 2) Cost of power regulation for each micro-source

[0195]

[0196] In the formula, P i fst (t) The power value of the i-th micro-source at time t obtained in the first stage, P i sn d(t) represents the power value of the i-th micro-source at time t in the second stage. The penalty coefficient is adjusted for the power value of the i-th micro-source in the second stage.

[0197] 3) Operating and maintenance costs of supercapacitors

[0198]

[0199] In the formula, K represents the planned output value of the supercapacitor at time t in the second stage. om.SC This represents the operating and maintenance cost coefficient for supercapacitors.

[0200] As a specific embodiment, the process of determining the intraday two-layer rolling optimization scheduling constraints in step S3 is as follows:

[0201] 1. Electric vehicle management:

[0202] SOC min ≤SOC n (t)≤SOC max

[0203]

[0204]

[0205] In the formula, This represents the charging and discharging power of the nth electric vehicle during time period t. These represent the charging and discharging flags of the nth electric vehicle during time period t.

[0206] 2. MPC Scrolling Optimization Layer

[0207] The economic control constraints are the same as those of the previous period.

[0208] As a specific embodiment, in step S4, the multi-timescale optimization model framework is as follows: First, in the day-ahead phase, economic dispatch is performed based on the obtained day-ahead forecasts of wind power, photovoltaic power output, and load power, and the results serve as reference values ​​for intraday rolling optimization; second, as... Figure 2As shown, the intraday phase is divided into two layers: the EV charging and discharging management layer and the MPC rolling optimization layer. The EV charging and discharging management layer optimizes the charging and discharging of EVs under various V2G modes based on the historical data of vehicle owners, obtaining the load data of the superimposed electric vehicles and inputting it into the MPC rolling optimization layer. The MPC rolling optimization layer is divided into two stages. The first stage is based on T... u For the time window, Δt u Economic scheduling is performed for a time step, and the result is used as a reference value in the second stage; the second stage is based on T. l For the time window, Δt l Using a time step, V2G load information is updated based on received real-time forecast data from wind and solar power, as well as the identification of EVs undergoing emergency charging and unplanned travel. This coordinates the output of distributed power sources, ensuring that the power connected to the main grid tracks the planned value from the upstream provider. When the second phase of forward rolling optimization is executed and Δt is reached... u After a certain period, return to the first stage, scroll the time window forward, and repeat the above process. The results of the first stage provide a reference for the control variables in the second stage, and the second stage feeds back the actual values ​​of the state variables to the first stage, serving as a feedback correction mechanism.

[0209] This invention proposes a two-layer rolling optimization model for intraday operation. When running in the MPC rolling optimization layer, in addition to using updated forecast data, it can also detect unplanned EV behavior, correct the V2G load output by the EV charging and discharging management layer, improve the economy and accuracy of scheduling, and enhance the tracking capability with the main network tie-line planned values.

[0210] While embodiments of the present invention have been given, those skilled in the art should understand that, based on the technical solutions of the present invention, modifications can be made to the embodiments herein without departing from the spirit of the invention. The above embodiments are merely illustrative and should not be construed as limiting the scope of the invention.

Claims

1. A multi-time-scale hierarchical scheduling method for cogeneration systems containing electric vehicles, characterized in that: S1. Establish a combined heat and power microgrid system; S2. Establish mathematical models for equipment in a combined heat and power microgrid system: S3. Determine the objective function and day-ahead economic scheduling constraints; S4. Establish a multi-timescale optimization model for cogeneration microgrids: The multi-timescale optimization model for cogeneration microgrids includes two timescales: Day-ahead: forecasting the output and load of new energy sources in the microgrid; Intra-day: joint optimization of the EV charging and discharging management layer and the Model Predictive Control (MPC) rolling optimization layer; The EV charging and discharging management layer takes into account the battery degradation costs caused by the charging and discharging of electric vehicles and establishes a multi-objective optimization scheduling model that minimizes the load curve variance and minimizes the dissatisfaction of EV owners participating in the V2G (Vehicle-to-Everything) mechanism; When running in the MPC rolling optimization layer, if the number of unplanned EVs exceeds the threshold, it will return to the EV charging and discharging management layer to solve again and then return. Step S3 is as follows: The day-ahead economic dispatch optimization cost function for the S3.1 microgrid operation is as follows: In the formula, This represents the cost of exchanging power with the main power grid. This represents the maintenance cost of distributed generation (DG). This indicates the operating cost of the gas turbine. Indicates the cost of treating polluting gases. This represents minimizing the day-ahead scheduling cost; S3.2 Determine the day-ahead economic dispatch constraints, including power balance constraints, distributed power output constraints, micro gas turbine ramping constraints, microgrid and main grid power exchange constraints, main grid power purchase and sale status mutual exclusion constraints, micro gas turbine start-up and shutdown time constraints, battery operation constraints, and thermal storage device constraints. S3.3 Determine the objective function for intraday two-level rolling optimization scheduling: Electric vehicle charging and discharging management layer: Objective function 1: In the formula, express t Load power value at any time This represents the average load power at all times. T For the scheduling period, This represents minimizing the variance of the total load curve; Objective function 2: In the formula, for t Time of the first n The degradation cost of an electric vehicle , These are the minimum and maximum degradation costs accepted by the user when signing the agreement. This represents the total number of electric vehicles. For the scheduling period, This indicates minimizing user dissatisfaction with participating in connected car initiatives; Model predictive control MPC rolling optimization layer: The first phase considers the highest economic efficiency when the combined heat and power microgrid is connected to the grid, and establishes an economic optimization model: In the formula, , , , These represent the costs of power exchange with the main grid, maintenance costs of distributed power sources, operating costs of micro gas turbines, and costs of treating pollutants in the first phase, respectively. and These are the reference values ​​for the power exchanged with the main grid and the power of the micro gas turbine at time t, obtained from the optimized scheduling a few days ago; and These represent the planned values ​​for the power exchanged with the main grid and the power of the micro gas turbine during the first stage at time t, respectively. and The penalty coefficients are respectively the power exchanged with the main grid in the first stage and the power of the micro gas turbine. This represents minimizing the total operating cost of the first phase; The optimization objective of the second stage is to balance the impact of power prediction errors based on the output of each unit obtained in the first stage. in , , These are the power regulation costs of the interconnection line with the main grid, the power regulation costs of the i-th distributed power source, and the operation and maintenance costs of the supercapacitor, respectively. This represents the total number of distributed power sources. This indicates minimizing the adjustment cost in the second stage; S3.4 Intraday two-layer rolling optimization constraints: The electric vehicle charging and discharging management layer includes the upper and lower limits of the electric vehicle charging and discharging power, the charging and discharging state variables of the electric vehicle at a certain moment, and the upper and lower limits of the electric vehicle battery capacity; The model predictive control MPC rolling optimization layer satisfies the constraints in S3.

2. The specific process of step S4 is as follows: S4.1 In the day-ahead phase: Economic dispatch is carried out based on the obtained day-ahead forecasts of wind power, photovoltaic power output and load power, and the results are used as reference values ​​for intraday rolling optimization; The S4.2 EV charging and discharging management layer optimizes the charging and discharging of EVs under various V2G modes based on the historical habit data of car owners, obtains the load data of superimposed electric vehicles, and inputs it into the model predictive control MPC rolling optimization layer. The first stage of the S4.3 MPC rolling optimization layer is based on For time window, Economic scheduling is performed for the time step, and the result is passed as a reference value to the second stage S4.4; The second stage of the S4.4 MPC rolling optimization layer is based on For time window, The V2G load information is updated based on the received real-time forecast data of wind and solar power and the identification of EVs that are charging urgently or traveling unplanned, so as to coordinate the output of each distributed power source and make the power connected to the main grid track the planned value of the upper level. S4.5 When the second stage forward roll optimization was executed After a certain period of time, return to the first stage, scroll the time window forward, and repeat the above process.

2. The multi-time-scale hierarchical scheduling method for a combined heat and power system containing electric vehicles as described in claim 1, characterized in that: In step S1, the equipment in the combined heat and power microgrid system includes micro gas turbine units, absorption chiller units, ground source heat pump units, wind turbine units, photovoltaic power generation modules, batteries and supercapacitors, and thermal storage devices.

3. The multi-time-scale hierarchical scheduling method for a combined heat and power system containing electric vehicles as described in claim 1, characterized in that, In step S3.2, the day-ahead economic dispatch constraints specifically include: Power balance constraints: In the formula, , , , , , They are respectively t The power consumption of the storage battery, the power consumption of the micro gas turbine, the power consumption of the photovoltaic module, the power consumption of the wind turbine generator, the power exchanged with the main grid, and the power consumption of the electrical load are all measured at all times. , , , They are respectively t The output heat power of the absorption chiller, the output heat power of the ground source heat pump unit, the output heat power of the thermal storage device, and the heat load power; Distributed power supply output power constraints In the formula, , , These are the output power of the i-th distributed power source and the lower and upper limits of the output power of the i-th distributed power source, respectively. Micro gas turbine ramping constraints In the formula, These represent the maximum rate of increase and rate of decrease of the gas turbine's output power. , These are the output power values ​​of the micro gas turbine at time t and t-1, respectively. For time step; Microgrids exchange power constraints with the main grid In the formula, Let be the absolute value of the power exchanged with the main network at time t. This is the upper limit of the power exchanged with the main network; Mutual Exclusion Constraints of Main Grid Power Purchase and Sale Status In the formula, for t Status of electricity purchase and sale on the main grid during the time period Micro gas turbine start-stop time constraints In the formula, This represents the start-up and shutdown state of the gas turbine during the time period t-k+1. These are the minimum start-up time and minimum shutdown time of the gas turbine, respectively. Battery operating constraints In the formula, and These correspond to the minimum and maximum allowable SOC values ​​of the battery during operation, respectively. The maximum allowable charging and discharging power of the battery; The maximum range of battery SOC change after one scheduling cycle; for t Battery charge / discharge status during a given period; The scheduling period; thermal storage device constraints In the formula: This refers to the rated capacity of the thermal storage device; , These are the minimum and maximum capacity coefficients of the thermal storage device, respectively. , These represent the maximum charging and discharging power of the thermal storage device.

4. The multi-time-scale hierarchical scheduling method for a combined heat and power system containing electric vehicles as described in claim 1, characterized in that, In step S3.4, the constraints on the upper and lower limits of the electric vehicle's charging and discharging power, the constraints on the electric vehicle's charging and discharging state variables at a certain moment, and the constraints on the upper and lower limits of the electric vehicle's battery capacity specifically include: In the formula, This indicates that the nth electric vehicle is in the time period t The charging and discharging power, , They represent the first n electric vehicles during the period t Charge and discharge indicators, for t Time of the first n The SOC value of an electric vehicle , These are the lower and upper limits of the State of Charge (SOC) for electric vehicles, respectively.

5. The multi-time-scale hierarchical scheduling method for a combined heat and power system containing electric vehicles as described in claim 1, characterized in that, Considering the impact of uncertainties in wind power, photovoltaics, and load forecasting on cogeneration microgrid systems, a beta distribution is used to fit the power forecasting errors of wind and photovoltaics, which can reasonably reflect the actual situation.

6. The multi-time-scale hierarchical scheduling method for a combined heat and power system containing electric vehicles as described in claim 1, characterized in that, In step S4, the CPLEX solver and a heuristic algorithm were used together to solve the multi-timescale optimization model of the cogeneration microgrid.

Images