Urban virtual power plant distribution robust optimization scheduling method and system

Abstract

The invention relates to an urban virtual power plant distribution robust optimization scheduling method and system, belongs to the technical field of power systems, and solves the problem that an existing robust optimization model is difficult to determine a proper uncertainty set. The method comprises the following steps: constructing internal output unit models of an urban virtual power plant, wherein the internal output unit models comprise an electrolytic hydrogen production model, a hydrogen storage system model, a fuel cell model, a waste incineration generator set output model and a flue gas purification device output model; constructing a target function and constraint conditions by taking the minimum operation cost, the minimum carbon emission and the maximum demand response benefit as targets; constructing an opportunity constraint distribution robustness model; the target function of multiple targets is converted into an optimal compromise single target function through normalization optimal compromise processing; and obtaining an optimal compromise solution, a corresponding confidence level and a scheduling scheme by using the multi-target optimal compromise economic low-carbon benefit optimization model. The minimization of the operation cost, the minimization of the carbon emission, the maximization of the demand response benefit and the maximization of the robustness are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power systems, and in particular to a distributed and optimized dispatching method and system for urban virtual power plants. Background Art

[0002] In order to cope with global climate change and energy crisis, distributed renewable energy (DRE) represented by wind and solar power has been connected to the distribution network in large quantities, injecting new impetus into the construction of new power systems. However, wind and solar power have large volatility and intermittency, and the phenomenon of wind and solar power abandonment is serious, which poses a challenge to the stable and efficient operation of new power systems. Virtual Power Plant (VPP) has become an important subject of DRE management due to its advanced distributed energy aggregation and control technology. It can not only consume the power of the power system in a planned way, but also output power to the power system in reverse, and carry out "peak shaving and valley filling" more flexibly and efficiently, and obtain considerable economic benefits, providing guarantee for the stability of the power system.

[0003] Research on VPP mainly focuses on the effective integration of distributed renewable energy, adjustable loads, microgrids, pumped storage, electric vehicles, etc., and their participation in power system operation and power market transactions, so as to achieve efficient utilization of distributed renewable energy and grid security and economic dispatch through the complementarity of aggregated resources. Research on the participation of hydrogen systems in VPP mainly builds VPPs of power-to-gas-gas-hydrogen blending, wind-solar-hydrogen-heat, and photovoltaic-coupled water electrolysis hydrogen production systems, studies their operating modes, and conducts research on their economic and feasibility. However, there is little analysis of its low-carbon nature. In addition, with the rapid development of urbanization, the output of urban garbage has increased at an annual rate of 8% to 10%, exacerbating urban environmental pollution.

[0004] In terms of research methods, VPPs are often distributed with a large number of distributed renewable energy sources, such as wind power and photovoltaics, and their uncertainty makes it difficult to dispatch and operate VPPs. In order to cope with the uncertainty of renewable energy, robust optimization methods are widely used because they do not require a large number of samples and can meet the operating constraints of all random scenarios. The robust optimization method is used to characterize the uncertainty of wind power in VPPs, and a robust bidding model for VPPs is proposed; the intraday scheduling problem of VPPs is studied based on robust optimization, and the example results show that this method can reduce the actual daily operating cost and improve the robustness of the optimization strategy. However, the above-mentioned traditional robust optimization model is difficult to determine a suitable uncertainty set. When the uncertainty set is too large, the scheduling strategy will be very conservative; otherwise, it will be too aggressive. Summary of the invention

[0005] In view of the above analysis, an embodiment of the present invention aims to provide a method and system for robust optimization scheduling of urban virtual power plants, so as to solve the problem that it is difficult to determine a suitable uncertainty set in the existing robust optimization model.

[0006] On the one hand, an embodiment of the present invention provides a method for distributed robust optimization scheduling of urban virtual power plants, including: constructing an internal output unit model of the urban virtual power plant, wherein the internal output unit model includes an electrolysis hydrogen production model, a hydrogen storage system model, a fuel cell model, a waste incineration power generation unit output model and a flue gas purification device output model; based on the internal output unit model, constructing an objective function and corresponding constraints with the objectives of minimizing the operating cost, minimizing carbon emissions, maximizing demand response benefits and maximizing robustness of the urban virtual power plant; constructing an opportunity constrained distributed robustness model based on a fuzzy set of wind and solar output prediction errors; converting a multi-objective objective function into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model; and using the multi-objective optimal compromise economic low-carbon benefit optimization model to obtain an optimal compromise solution, a corresponding confidence level and a scheduling plan.

[0007] The beneficial effects of the above technical solution are as follows: According to the distributed robust optimization scheduling method of urban virtual power plants in the embodiment, in order to solve the urban garbage problem, a garbage power generation system is introduced, which is combined with distributed power sources such as gas turbines, wind and solar power, and flexible loads and hydrogen energy storage systems to form a new type of urban virtual power plant. For the uncertainty of wind and solar power that needs to be considered in the optimization scheduling of virtual power plants, an economic-low-carbon-benefit-robust multi-objective collaborative optimization scheduling model under multi-dimensional uncertain factors is established to eliminate the limitations of the subjective setting of robustness in conventional distributed robust optimization, and ultimately achieve the multiple goals of minimizing operating costs, minimizing carbon emissions, maximizing demand response benefits and maximizing robustness.

[0008] Based on the further improvement of the above method, constructing the objective function based on the internal output unit model with the goal of minimizing the operating cost of the town virtual power plant, minimizing carbon emissions, maximizing demand response benefits and maximizing robustness further includes: expressing the first objective function of minimizing the operating cost of the town virtual power plant by the following formula:

[0009] minF1=C(P g )+C OP +C c ;

[0010]

[0011] Among them, C(P g ) represents the power generation cost of the micro gas turbine; C OPrepresents the operation and maintenance costs of wind power, photovoltaic electrolyzers, fuel cells, waste incineration generators and flue gas purification devices; C c represents the carbon trading cost; C i (P g,i,t ) is the cost function of micro gas turbine, P g,i,t is the power generation of the i-th micro gas turbine at time t; a i , b i 、c i are the power generation cost coefficients of the i-th micro gas turbine, N is the total number of micro gas turbines; β WT , β PV , β ele , β fc , β WI , β WI,s Respectively represent the operating cost coefficients of wind power, photovoltaic, electrolyzer, fuel cell, waste incineration generator set, and flue gas purification device; β WT,t , P PV,t , P ele,n,t , P fc,t , P WI (t), P WI,s (t) are the output of wind power, photovoltaic power, electrolyzer, fuel cell, waste incineration generator set and flue gas purification device at time t obtained based on the models of each internal output unit; λ c,t is the price of unit carbon emission rights at time t; ΔE represents the difference in carbon emissions; k i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; γ is the carbon emission quota allocation coefficient of the urban virtual power plant; P WT,t represents the wind power at time t; P PV,t represents the photovoltaic power generation at time t; the second objective function of the lowest carbon emission of the urban virtual power plant is expressed by the following formula:

[0012]

[0013] Among them, P g,i,t is the power generation of the i-th micro gas turbine at time t; κ i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; P WI,t is the power generation of the i-th waste incineration power generation unit at time t; κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; the third objective function of maximizing the demand response benefit of the urban virtual power plant is expressed by the following formula:

[0014]

[0015] Where U is the benefit of demand response, x i,t The power reduction of the i-th type of demand response user at time t, λ i x i,t represents the user response x of the i-th demand response at time t i,t When the power consumption is less, the marginal benefit of the urban virtual power plant due to less power generation; n i,t represents the subsidy paid to the i-th type of demand response user at time t, m 1,i 、m 2,i is the incentive cost coefficient of the i-th type of demand response user, θ i is the response willingness of the i-th type of demand response user; and

[0016] The fourth objective function with the highest robustness is expressed by the following formula:

[0017]

[0018] in, is the cumulative probability corresponding to the upper and lower bounds of the robust operating interval.

[0019] Based on the further improvement of the above method, the objective function of multiple objectives is converted into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model further comprising: normalizing the first objective function, the second objective function, the third objective function and the fourth objective function through the following formula:

[0020]

[0021] Among them, F represents the vector form of system economy, low carbon and demand response benefits; F i represents the i-th target; are the minimum and maximum values ​​of the i-th target, respectively, Z i Represents the normalized value of the i-th target; according to the above normalization processing formula, the normalized optimal solution and the worst solution are obtained, so as to calculate the relative distance of the scheduling scheme Z from the optimal solution and the worst solution by the following formula:

[0022]

[0023] Among them, D + , D - Respectively represent the relative distances of the system scheduling solution from the optimal solution and the worst solution; based on the relative distances of the system scheduling solution from the optimal solution and the worst solution, the following multi-objective optimal compromise function is constructed:

[0024]

[0025] Wherein, W is the compromise optimization degree, 0≤W≤1, and the merits of the scheme are judged according to the compromise optimization degree.

[0026] Based on the further improvement of the above method, the constraints include: load supply and demand balance constraint, hydrogen energy balance constraint, micro gas turbine output constraint, wind and solar output constraint, electrolyzer constraint, hydrogen storage tank constraint, fuel cell power constraint, waste incineration generator power constraint, flue gas purification device power constraint and demand response constraint, wherein the load supply and demand balance constraint is that the sum of the power output power of the urban virtual power plant at time t meets all power demands at time t; the hydrogen energy balance constraint is that at each time point all the hydrogen produced is equal to all the hydrogen consumed to ensure the balance of hydrogen supply and demand; the micro gas turbine output constraint is to ensure that the micro gas turbine operates within a safe and economic range by setting the output range of each micro gas turbine and the output change rate in adjacent time intervals; the wind and solar output constraint is that the output of wind power and photovoltaic power generation is set between 0 and its maximum output to ensure that wind turbines and photovoltaic generators The group operates within a safe and economical range; the electrolyzer constraint is to set the power constraint of the electrolyzer and the start and stop constraint of the electrolyzer to ensure the safe and reliable operation of the electrolyzer; the hydrogen storage tank constraint is to set the upper and lower limit constraints and continuity constraints of the hydrogen storage tank; the fuel cell power constraint is to set the fuel cell power between 0 and the maximum power of the fuel cell; the waste incineration power generation group power constraint is to set the waste incineration power generation group power between 0 and the maximum power of the waste incineration power generation group; the flue gas purification device power constraint is to set the flue gas purification device power between 0 and the maximum power of the flue gas purification device; the waste resource utilization potential constraint is to reasonably arrange the operation plan of the waste incineration power generation group according to the change of the utilization potential of the waste resources at time t during the scheduling process to ensure that its output is within the system demand range; and the demand response constraint is to set the load reduction of each type of user demand response within a reasonable range.

[0027] Based on the further improvement of the above method, the load supply and demand balance constraint is calculated by the following formula:

[0028] P WT,t +P PV,t +P g,t +P fc,t +P WI,t =L t +P ele,i,t +P WI,s,t ;

[0029] Among them, P WT,t represents the wind power at time t; P PV,t represents the photovoltaic power generation at time t, P g,t represents the power of the micro gas turbine at time t, Pfc,t represents the power of the fuel cell at time t, P WI,t Indicates the power of the waste incineration generator set at time t; L t Indicates the power load, P ele,i,t represents the power of the electrolytic cell at time t, P WI,s,t represents the power of the flue gas purification device at time t; the hydrogen energy balance constraint is calculated by the following formula:

[0030] M ele,t +M loss,t =L h,t +M sto,t +M fc,t ;

[0031]

[0032] Among them, M ele,t represents the mass of hydrogen produced by the electrolyzer at time t; M loss,t represents the mass of hydrogen lost at time t; L h,t represents the hydrogen load at time t; M sto,t Represents the mass of hydrogen absorbed or released by the hydrogen storage tank at time t, with absorption being positive and release being negative; M fc,t represents the mass of hydrogen consumed by the fuel cell at time t; η H2E The efficiency of the electrolyzer; HHV represents the calorific value of hydrogen, which is 39.44 kW·h / kg; the output constraint of the micro gas turbine is calculated by the following formula:

[0033]

[0034] P g,i,min ≤P g,i,t ≤P g,i,max ;

[0035] P d,i ≤P g,i,t+1 -P g,i,t ≤P u,i ;

[0036] Among them, P g,t represents the power of the micro gas turbine at time t; P g,i,t represents the power of the i-th micro gas turbine at time t; P g,i,min , P g,i,max Respectively represent the lower and upper limits of the output of the i-th micro gas turbine; P d,i , P u,i They represent the downhill climbing and uphill climbing of the i-th micro gas turbine respectively; the wind-solar output constraint is calculated by the following formula:

[0037] 0≤P WT,t ≤P WT,max;

[0038] 0≤P PV,t ≤P PV,max ;

[0039] Among them, P WT,max , P PV,max are the maximum outputs of wind power and photovoltaic power, respectively; the power constraint and the start-stop constraint of the electrolyzer are calculated by the following formula:

[0040] P ele,min U ele,i,t ≤P ele,i,t ≤P ele,max U ele,i,t ;

[0041]

[0042] Among them, P ele,min and P ele,max are the minimum and maximum power of the electrolyzer, in kW, and the minimum and maximum power of the electrolyzer are set to 25% and 100% of the rated power respectively; U ele,i,t is the working status flag of the i-th electrolytic cell at time t, 1 indicates operation, 0 indicates shutdown; Y i,t , Z i,t represents the start and stop state of the i-th electrolyzer at time t; ON represents the maximum number of starts allowed per day for the electrolyzer, and OFF represents the maximum number of stops allowed per day for the electrolyzer; the upper and lower limit constraints and continuity constraints of the hydrogen storage tank are calculated by the following formula:

[0043] SOHC min ≤SOHC t ≤SOHC max ;

[0044]

[0045] Among them, SOHC min and SOHC max They are the minimum and maximum thresholds of the hydrogen storage state of the hydrogen storage tank; SOHC t is the hydrogen storage state of the hydrogen storage tank at time t; H max Represents the total capacity of the hydrogen storage tank, in kg; SOHC init and SOHC final Respectively represent the SOHC value of the hydrogen storage tank at the initial time and the final time of the day; the upper and lower limits of the fuel cell power are calculated by the following formula:

[0046] 0≤P fc,t ≤P fc,max ;

[0047] Among them, P fc,max represents the maximum power of the fuel cell; the power constraint of the waste incineration power generation unit is calculated by the following formula:

[0048] 0≤P WI,t ≤P WI,max ;

[0049] Among them, P WI,max represents the maximum power of the waste incineration generator set; the power constraint of the flue gas purification device is calculated by the following formula:

[0050] 0≤P WI,s,t ≤P WI,s,max

[0051] Among them, P WI,s,max represents the maximum power of the flue gas purification device; the waste resource utilization potential constraint is calculated by the following formula:

[0052]

[0053] in, They represent the minimum and maximum supply of garbage resources in the cycle respectively; are the minimum combustible waste ratio and the maximum combustible waste ratio respectively; the demand response constraint is calculated by the following formula:

[0054]

[0055] x i,min ≤x i,t ≤x i,max ;

[0056]

[0057] Among them, D t represents the sum of load reductions in response to demand by all user types at time t; x i,t represents the load reduction of the i-th user at time t; x i,min 、x i,max They represent the minimum reduction and maximum reduction at any time point t; M represents the total number of user types; D L,i It represents the maximum total load reduction of the i-th user in response to demand reduction on a daily basis; The value range of user type i;

[0058]

[0059] Among them, n i,t is the subsidy paid by the urban virtual power plant to the i-th user at time t, It represents the outage cost of the i-th type of demand response user.

[0060] Based on the further improvement of the above method, the internal output unit model also includes a gas turbine model, a wind power output model and a photovoltaic output model, wherein the gas turbine model is represented by the following formula:

[0061]

[0062] P g,min ≤P g ≤P g,max ;

[0063] P d ≤P g,t+1 -P g,t ≤P u ;

[0064] Among them, C g represents the power generation cost of micro gas turbine, a, b, c are power generation cost coefficients, P g is the power generation power of the micro gas turbine; g,min , P g,max Respectively represent the lower and upper limits of gas turbine output, P g,t , P g,t+1 Respectively represent the output power at time point t and the next time point t+1, P d , P u They represent the maximum down-slope rate and up-slope rate acceptable to the gas turbine respectively; the wind power output model is represented by the following formula:

[0065]

[0066] Among them, v ci is the cut-in wind speed; v co is the cut-out wind speed; v r is the rated wind speed; v is the natural wind speed, P WT Indicates the active output power of the fan; P WT,r Indicates the rated active power of the wind turbine; the photovoltaic output model is expressed by the following formula:

[0067]

[0068] Among them, R c Indicates the actual solar radiation intensity, R max Indicates the maximum solar radiation intensity, P PV Represents the photovoltaic active output power, P max 、T max At 1000W / m 2 , maximum output power and maximum reference temperature under 25℃ standard conditions, Tc represents the actual operating temperature, k represents the power temperature coefficient; the electrolysis hydrogen production model is represented by the following formula:

[0069] M ele =0.017P ele +0.22;

[0070] Among them, M ele Indicates the mass of hydrogen produced per hour, in kg; P ele is the electrolyzer power; the hydrogen storage state SOHC of the hydrogen storage tank in the hydrogen storage system model is expressed by the following formula:

[0071]

[0072] M sto,t =M ele,t -M fe,t -L h,t ;

[0073] Among them, M sto,t Indicates the mass of hydrogen filled or discharged in the hydrogen storage tank at time t, in kg, M sto,t A positive value indicates that the hydrogen storage tank is filled with hydrogen at time t, M sto,t Negative indicates that the hydrogen storage tank releases hydrogen at time t, H max It represents the capacity of the hydrogen storage tank in kg; Δt represents a time interval; M ele,t represents the mass of hydrogen produced by the electrolyzer at time t, L h,t represents the hydrogen load at time t, M fe,t Indicates the mass of hydrogen consumed by the fuel cell for power generation at time t, which is always 0 in the grid-connected scenario; the power generation power of the proton exchange membrane fuel cell PEMFC in the fuel cell model is expressed by the following formula:

[0074]

[0075] Among them, P fc (t) represents the power generation of the fuel cell at time t, V H (t) represents the hydrogen consumption of the fuel cell at time t, Q H Indicates the calorific value of hydrogen, η fc-e The efficiency of fuel cell power generation is represented by the following formula:

[0076] P WI (t) = λ WI ·R WI (t);

[0077] P WI,n (t) = P WI (t)-P WI，s(t);

[0078] Among them, P WI (t) represents the power generation of waste incineration at time t; WI Indicates the energy conversion coefficient of combustible waste. The electricity converted per ton of domestic waste entering the factory is 280Wh / t; R WI (t) represents the amount of combustible waste burned (kg); P WI，n (t) is the net power generation of the waste incineration power generation system; P WI，s (t) represents the total power consumption of the flue gas at time t kW; the output model of the flue gas purification device is represented by the following formula:

[0079]

[0080] F sk (t) = P WI (t)·e WI =F sk,cl (t)+F sk,s (t);

[0081]

[0082] Where, ρ represents the smoke split ratio (%); F sk (t) represents the total flue gas volume generated by waste incineration power generation at time t; F sk,cl (t) represents the amount of flue gas directly treated at time t; F sk,s (t) represents the amount of smoke entering the smoke storage tank at time t; e WI It represents the flue gas emission intensity per unit of waste incineration power generation system (t / kW·h); P WI (t) represents the power generation of waste incineration at time t kW; P WI,s (t) represents the total power consumption of flue gas treatment at time t kW; μ WI,s Indicates the unit energy consumption coefficient of flue gas treatment; μ GP Indicates the energy consumption of the air pump when the gas enters the flue gas; It indicates the amount of smoke in the smoke storage tank entering the reaction tower at time t.

[0083] Based on the further improvement of the above method, the opportunity constraint distribution robustness model based on the fuzzy set of wind and solar output forecast error is further constructed including:

[0084] At the confidence level γ, the true probability θ interval is as follows:

[0085]

[0086] Among them, H and G are the Beta distribution B(m i ,s+nmi ) and Beta distribution B(s+m i ,nm i )’s cumulative distribution function; θ1 and θ′ are the upper and lower bounds of the true probability distribution of θ; m i and n represent the number of samples and the total number of samples where the random variable ξ is less than sample A respectively; the fuzzy set of wind and solar power output prediction error is expressed as:

[0087] P={F∈P0(|ξ1,ξ′|)|F[ξ<ξ i ]∈[θ 1i <θ i ′],i=1,2,…,n};

[0088] Among them, P0(|ξ1,ξ′|) represents the set of all probabilities within the estimation range |ξ1,ξ′|; F is the sample point ξ i The cumulative probability distribution of ; Based on the fuzzy set of wind and solar power output prediction error, the following opportunity constraint distribution robustness model is constructed:

[0089]

[0090] Among them, Ψ is the robustness of a scheduling cycle; Γ t is the robustness corresponding to the robust operation interval at time t; R is the robust interval set; R t are the robust operating interval and interval variable respectively; are the upper and lower bounds of the interval respectively; R r Indicates that the time point t falls into the robust interval The probability of is the cumulative probability corresponding to the upper and lower bounds of the robust operation interval; F WP,t (·) is the cumulative probability distribution function of wind power forecast error at time t; Provide upper and lower spare capacity for coal-fired units at time t; K w is the wind power installed capacity; the opportunity constraint distribution robustness model under the joint probability distribution of wind and solar forecast errors is constructed as follows:

[0091]

[0092] Among them, F 1,t (·), F′ t (·) is the upper and lower bounds of the joint cumulative probability distribution of the wind-solar joint prediction error at time t.

[0093] On the other hand, an embodiment of the present invention provides a distributed robust optimization scheduling system for urban virtual power plants, including: an output unit model construction module, used to construct an internal output unit model of the urban virtual power plant, wherein the internal output unit model includes an electrolysis hydrogen production model, a hydrogen storage system model, a fuel cell model, a waste incineration power generation unit output model and a flue gas purification device output model; an objective function construction module, used to construct an objective function and corresponding constraints based on the internal output unit model with the objectives of minimizing the operating cost, minimizing carbon emissions, maximizing demand response benefits and maximizing robustness of the urban virtual power plant; a robustness model construction module, used to construct an opportunity-constrained distributed robustness model based on a fuzzy set of wind and solar output prediction errors; and a conversion module, used to convert a multi-objective objective function into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model; a solution module, used to use the multi-objective optimal compromise economic low-carbon benefit optimization model to obtain an optimal compromise solution, a corresponding confidence level and a scheduling plan.

[0094] Based on the further improvement of the above system, the objective function construction module is also used to construct the first objective function, the second objective function, the third objective function and the fourth objective function, wherein the first objective function with the minimum operating cost of the town virtual power plant is expressed by the following formula:

[0095] minF1=C(P g )+C OP +C c ;

[0096]

[0097]

[0098] Among them, C(P g ) represents the power generation cost of the micro gas turbine; C OP represents the operation and maintenance costs of wind power, photovoltaic electrolyzers, fuel cells, waste incineration generators and flue gas purification devices; C c represents the carbon trading cost; C i (P g,i,t ) is the cost function of micro gas turbine, P g,i,t is the power generation of the i-th micro gas turbine at time t; a i , b i 、c i are the power generation cost coefficients of the i-th micro gas turbine, N is the total number of micro gas turbines; β WT , β PV , β ele , β fc , β WI , βWI,s Respectively represent the operating cost coefficients of wind power, photovoltaic, electrolyzer, fuel cell, waste incineration generator set, and flue gas purification device; P WT,t , P PV,t , P ele,n,t , P fc,t , P WI (t), P WI,s (t) are the output of wind power, photovoltaic power, electrolyzer, fuel cell, waste incineration generator set and flue gas purification device at time t obtained based on the models of each internal output unit; λ c,t is the price of unit carbon emission rights at time t; ΔE represents the difference in carbon emissions; k i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; γ is the carbon emission quota allocation coefficient of the urban virtual power plant; P WT,t represents the wind power at time t; P PV,t represents the photovoltaic power generation at time t; the second objective function of the lowest carbon emission of the urban virtual power plant is represented by the following formula:

[0099]

[0100] Among them, P g,i,t is the power generation of the i-th micro gas turbine at time t; κ i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; P WI,t is the power generation of the i-th waste incineration power generation unit at time t; κ i2 is the carbon emission intensity per unit electricity of the i-th waste incineration power generation unit; the third objective function with the maximum demand response benefit of the urban virtual power plant is expressed by the following formula:

[0101]

[0102] Where U is the benefit of demand response, x i,t The power reduction of the i-th type of demand response user at time t, λ i x i,t represents the user response x of the i-th demand response at time t i,t When the power consumption is less, the marginal benefit of the urban virtual power plant due to less power generation; n i,t represents the subsidy paid to the i-th type of demand response user at time t, m 1,i 、m 2,i is the incentive cost coefficient of the i-th type of demand response user, θ i is the response willingness of the i-th type of demand response user; and

[0103] The fourth objective function with the highest robustness is expressed by the following formula:

[0104]

[0105] in, is the cumulative probability corresponding to the upper and lower bounds of the robust operating interval.

[0106] Based on the further improvement of the above system, the conversion module is further used for: a normalization processing module, a relative distance determination module and a multi-objective optimal compromise function construction module, wherein the normalization processing module is used to normalize the first objective function, the second objective function, the third objective function and the fourth objective function by the following formula:

[0107]

[0108] Among them, F represents the vector form of system economy, low carbon and demand response benefits; F i represents the i-th target; are the minimum and maximum values ​​of the i-th target, respectively, Z i represents the normalized value of the i-th target; the relative distance determination module is used to obtain the normalized optimal solution and the worst solution according to the above normalization processing formula, so as to calculate the relative distance between the optimal solution and the worst solution of the scheduling scheme Z by the following formula:

[0109]

[0110] Among them, D + , D - Respectively represent the relative distances of the system scheduling solution from the optimal solution and the worst solution; the multi-objective optimal compromise function construction module is used to construct the following multi-objective optimal compromise function based on the relative distances of the system scheduling solution from the optimal solution and the worst solution:

[0111]

[0112] Wherein, W is the compromise optimization degree, 0≤W≤1, and the merits of the scheme are judged according to the compromise optimization degree.

[0113] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0114] 1. The technical solution of the present invention is based on a hydrogen energy storage town virtual power plant. By introducing waste incineration units to generate electricity from waste, it not only solves the problem of waste pollution of the environment, but also reduces the cost of waste disposal, thereby promoting regional carbon emission reduction and improving energy utilization efficiency.

[0115] 2. As an important part of virtual power plants, energy storage can make up for the defects of random volatility of renewable energy such as wind and solar power, and improve the flexibility of the power system. Hydrogen energy storage, as a clean and efficient energy source, can play an important role in VPP. Different from other energy storage systems, hydrogen energy storage systems can produce hydrogen through excess water electrolysis when there is excess renewable energy, and relieve load pressure by generating electricity through fuel cells during peak load periods. This can not only fully absorb resources such as abandoned wind, abandoned solar, and abandoned water, but also serve as energy storage to ensure power supply, enhance the added value within the system, and enhance the ability of VPP to participate in external markets.

[0116] 3. Based on multi-objective decision-making theory and approximate ideal solution sorting technology, a multi-objective optimal compromise function that comprehensively considers the economy, low carbon and demand response benefits of system scheduling is constructed, thereby establishing a multi-objective optimal compromise distributed robust scheduling model for hydrogen energy storage. This can quickly and accurately obtain the optimal scheduling plan for hydrogen energy storage urban virtual power plants that are both economical, low carbon and efficient, better guide the operation and trading of hydrogen energy storage virtual power plants, improve carbon emission reduction and energy utilization efficiency, and thus promote the construction of my country's new power system.

[0117] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0118] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0119] Figure 1 It is a flow chart of the method for optimizing the scheduling of urban virtual power plants according to this embodiment;

[0120] Figure 2 is a structural diagram of the urban virtual power plant operation system according to this embodiment;

[0121] Figure 3 is the active power output curve of wind power generation according to this embodiment;

[0122] Figure 4 is a structural diagram of the electrolysis hydrogen production system according to this embodiment;

[0123] Figure 5 is a flow chart of model solution according to this embodiment;

[0124] Figure 6A block diagram of a distributed optimization scheduling system for urban virtual power plants according to another embodiment. DETAILED DESCRIPTION

[0125] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0126] This application aims at the urban garbage problem by introducing a garbage power generation system, combining it with distributed power sources such as gas turbines, wind and solar power, flexible loads, and hydrogen energy storage systems to form a new type of urban virtual power plant. For the uncertainty of wind and solar power that needs to be considered when optimizing the scheduling of virtual power plants, an economic-low-carbon-benefit-robust multi-objective collaborative optimization scheduling model is established under multi-dimensional uncertain factors to eliminate the limitations of the subjective setting of robustness in conventional distributed robust optimization, and ultimately achieve the multiple goals of minimizing operating costs, minimizing carbon emissions, maximizing demand response benefits, and maximizing robustness.

[0127] The steps of the distributed robust optimization dispatching method of urban virtual power plants are as follows: a. description of VPP structure and internal output unit model; b. modeling of virtual power plant optimization dispatching; c. construction of distributed robust optimization model; d. solving the multi-objective optimal compromise economic low-carbon benefit robust model. The distributed robust optimization dispatching method of urban virtual power plants is described in detail below.

[0128] refer to Figure 1 A specific embodiment of the present invention discloses a method for optimizing the scheduling of urban virtual power plants, including:

[0129] In step S101, an internal output unit model of the urban virtual power plant is constructed, wherein the internal output unit model includes a gas turbine model, a wind power output model, a photovoltaic output model, an electrolysis hydrogen production model, a hydrogen storage system model, a fuel cell model, a waste incineration power generation unit output model and a flue gas purification device output model.

[0130] With the construction of new towns and the rapid development of urban economy, the amount of domestic waste has increased sharply, increasing the cost of waste disposal and the pollution to the environment. In response to the urban waste problem, by introducing waste power generation system, it is combined with gas turbines, wind and solar power and other distributed power sources and flexible loads, hydrogen energy storage system, to form a new type of urban virtual power plant, and realize profits by participating in the electricity market and carbon trading market. Figure 1 It is the VPP operation system.

[0131] 1. Gas turbine model: Micro gas turbine is a typical controllable DG with small size, simple structure, low carbon emissions, and power generation capacity usually ranging from several hundred kW to several hundred MW. It is an important aggregation unit of VPP. Gas turbine generates electricity by burning fossil fuels. It has strong climbing ability and short start-stop time. It is often used as a coordinated dispatching unit for wind and solar units with strong randomness, which can effectively reduce the impact of wind and solar fluctuations on the power grid. Adding a highly controllable micro gas turbine to VPP can improve the dispatchability of VPP.

[0132] The power generation cost function of a micro gas turbine is usually expressed as a quadratic function.

[0133]

[0134] Where: C g represents the power generation cost of micro gas turbine, a, b, c are power generation cost coefficients, P g is the power generation capacity of the micro gas turbine.

[0135] In addition, the gas turbine must also meet upper and lower limit constraints and climbing constraints.

[0136] P g,min ≤P g ≤P g,max (2)

[0137] P d ≤P g,t+1 -P g,t ≤P u (3)

[0138] Where: P g,min , P g,max Respectively represent the lower and upper limits of gas turbine output, P g,t , P g,t+1 Respectively represent the output power at time point t and the next time point t+1, P d , P u They represent the maximum acceptable downward ramp rate and upward ramp rate of the gas turbine respectively.

[0139] 2. Wind power output model: Wind energy is one of the most widely used clean energy sources. It has the advantages of large reserves, low cost, and wide distribution. It can generate electricity both during the day and at night. Wind power mainly relies on the wind to drive the blades of wind turbines to rotate for power generation. However, wind speed has strong randomness. Therefore, the Weibull distribution probability density function is often used to describe the wind speed model (refer to Figure 3 ).

[0140]

[0141] Where: v is the natural wind speed, k is the shape parameter, and c is the scale parameter.

[0142] The relationship between wind turbine output and wind speed is usually expressed as the cut-in wind speed v ci , Cut-out wind speed v co and rated wind speed v r Describe the relevant functions.

[0143]

[0144] Where: P WT Indicates the active output power of the fan; P WT,r Indicates the rated active power of the wind turbine. The active output power P of the wind turbine WT The approximate relationship with wind speed v is as follows Figure 2 shown.

[0145] 3. Photovoltaic output model: Photovoltaic power generation is the conversion of solar energy into electrical energy. Solar energy is one of the common clean energy sources. Its output power is mainly related to the intensity of solar radiation and ambient temperature. The light intensity is often described by the Beta distribution function:

[0146]

[0147] Where: τ(·) represents the Gamma function, R c Indicates the actual solar radiation intensity, R max represents the maximum solar radiation intensity, and α and β are the shape parameters of the Beta distribution function.

[0148] The functional relationship between photovoltaic output power, light intensity and ambient temperature is as follows:

[0149]

[0150] Where P PV Represents the photovoltaic active output power, P max 、T max At 1000W / m 2 , maximum output power and maximum reference temperature under 25℃ standard conditions, T c represents the actual operating temperature, and k represents the power temperature coefficient.

[0151] 4. Electrolysis hydrogen production model: There are two main methods of hydrogen production. One is to directly use fossil energy such as natural gas and methane and steam reforming to produce hydrogen, and the other is to electrolyze water to produce hydrogen. Compared with the use of fossil energy and steam reforming to produce hydrogen, there are problems of environmental pollution and low energy conversion efficiency. Electrolysis of water to produce hydrogen has the advantages of not consuming fossil energy and being clean and efficient. According to the different electrolyte materials, the commonly used electrolyzers can be divided into three types: alkaline electrolyzer (AEC), solid oxide electrolyzer (SOEC) and proton exchange membrane electrolyzer (PEMEC).

[0152] Compared with proton exchange membrane electrolyzers and solid oxide electrolyzers, which have high operating costs and immature technology, alkaline electrolyzers have become the most widely used electrolyzers due to their mature technology, low cost, and simple structure. This paper selects alkaline electrolyzers as the main equipment for hydrogen production by electrolyzing water.

[0153] The alkaline electrolytic cell hydrogen production equipment consists of alkaline electrolytic solution, DC power supply, water supply equipment, electrolytic electrodes, gas separation and compression equipment and hydrogen storage equipment. The structure of the electrolytic hydrogen production system is as follows Figure 4 shown.

[0154] According to the structure of the electrolytic hydrogen production system, the chemical reaction in the electrolytic cell is analyzed, and direct current is applied to the electrolytic cell containing water and electrolyte. An electrochemical reaction occurs, hydrogen is produced at the cathode, and oxygen is produced at the anode. The process of water electrolysis strictly abides by the chemical reaction equation, and according to the law of conservation of matter, the amount of hydrogen produced by the water electrolysis reaction is proportional to the size of the direct current.

[0155] The chemical reaction equation is as follows:

[0156] The overall reaction equation is:

[0157] 2H2O=2H2+O2

[0158] Under alkaline conditions:

[0159] cathode:

[0160] 4H2O+4e - =2H2+4OH -

[0161] anode:

[0162] 4OH - -4e - =2H20+O2

[0163] The first-order linear expression of the hydrogen mass produced per hour by the electrolyzer and the electrolyzer power is:

[0164] M ele =0.017P ele +0.22 (8)

[0165] M ele Indicates the mass of hydrogen produced per hour, in kg; P ele is the electrolytic cell power.

[0166] 5. Hydrogen storage system model: The hydrogen storage tank is a device for storing hydrogen. The hydrogen produced by electrolysis in the alkaline electrolyzer is stored in the hydrogen storage tank for use in fuel cell power generation and hydrogen load. There are three main ways to store hydrogen: compressed gas, cryogenic liquid, and solid. Among them, gaseous hydrogen storage is widely used due to its high conversion efficiency and low energy loss rate. Since the amount of hydrogen produced by electrolysis is expressed in mass, in order to correspond to it and facilitate calculation, the capacity of the hydrogen storage tank is also expressed in mass.

[0167] Hydrogen storage tanks can smooth out the fluctuation of hydrogen energy by filling and discharging hydrogen. In order to reflect the hydrogen storage state of the hydrogen storage tank, the hydrogen storage state SOHC (State of Hydrogen Charge) of the hydrogen storage tank is introduced.

[0168]

[0169] Where: M sto,t It indicates the mass of hydrogen filled or discharged in the hydrogen storage tank at time t, in kg, M sto,t A positive value indicates that the hydrogen storage tank is full of hydrogen at time t, M sto,t A negative value indicates that the hydrogen storage tank is releasing hydrogen at time t, H max It represents the capacity of the hydrogen storage tank in kg; Δt represents a time interval.

[0170] The mass of hydrogen charged and discharged from the hydrogen storage tank at time t is expressed by the following formula:

[0171] M sto,t =M ele,t -M fe,t -L h,t (10)

[0172] Where: M ele,t represents the mass of hydrogen produced by the electrolyzer at time t, L h,t represents the hydrogen load at time t, M fe,t Indicates the mass of hydrogen consumed by the fuel cell for power generation at time t, which is always 0 in the grid-connected scenario.

[0173] 6. Fuel cell model: A fuel cell is a device that converts the chemical energy in a fuel into electrical energy. When a fuel cell is working, fuel, such as hydrogen, is provided from the outside, and then a combustion reaction is carried out to convert the chemical energy therein into electrical energy. The internal reaction of a fuel cell is actually the reverse reaction of water electrolysis, that is, hydrogen and oxygen react to produce water. Fuel cells can be divided into two categories according to the temperature of the dielectric: high temperature (dielectric operating temperature 600°C and above) and low temperature (dielectric operating temperature 200°C and below). High temperature fuel cells can be divided into solid oxide fuel cells and molten carbonate fuel cells, and low temperature fuel cells can be divided into alkaline fuel cells, phosphoric acid fuel cells, and proton exchange membrane fuel cells.

[0174] In actual engineering, proton exchange membrane fuel cells (PEMFC) are widely used, so this article uses this fuel cell. Its chemical reaction formula is as follows:

[0175] anode:

[0176] H2=2H + +2e -

[0177] cathode:

[0178]

[0179] Overall reaction:

[0180]

[0181] The mathematical model of PEMFC power generation is as follows:

[0182]

[0183] Where: P fc (t) represents the power generation of the fuel cell at time t, V H (t) represents the hydrogen consumption of the fuel cell at time t, Q H Indicates the calorific value of hydrogen, η fc-e Indicates the efficiency of fuel cell power generation.

[0184] 7. Output model of waste incineration power generation unit: Waste incineration power generation is a single-input multi-output processing system, and its power generation is related to the power generation link and the flue gas purification treatment link. The article adopts the waste incineration power generation model with a decoupled flue gas purification device.

[0185] The net output model of the incineration generator set is:

[0186] P WI (t) = λ WI ·R WI (t) (12)

[0187] Where: P WI (t) represents the power generation of waste incineration at time t; WI Indicates the energy conversion coefficient of combustible waste. According to relevant national regulations, the electricity conversion rate for each ton of domestic waste entering the factory is 280Wh / t; P WI (t) represents the incineration amount of combustible garbage in the general education class (kg).

[0188] Taking into account the power consumed by the flue gas purification device, the net power generation of the waste incineration power generation system is P WI，n (t) is:

[0189] PWI,n (t) = P WI (t)-P WI，s (t) (13)

[0190] Where: P WI，s (t) represents the total power consumption of flue gas at time t (kW).

[0191] 8. Flue gas purification device output model: Flue gas treatment is generally carried out simultaneously with waste incineration power generation. However, by installing a smoke storage device, the power generation link and the flue gas purification link can be decoupled, so that the flue gas purification load can be transferred in different time periods. The flue gas generated by waste incineration enters the smoke storage tank or reaction tower. According to the peak and valley moments of the load, the flue gas diversion ratio is adjusted to adjust the flue gas entering the smoke storage tank, so that the energy consumption of the flue gas treatment device is adjustable. The specific calculation formula is as follows:

[0192]

[0193] F sk (t) = P WI (t)·e WI =F sk,cl (t)+F sk,s (t) (15)

[0194] Where: ρ represents the flue gas split ratio (%); F sk (t) represents the total flue gas volume generated by waste incineration power generation at time t; F sk,cl (t) represents the amount of flue gas directly treated at time t; F sk,s (t) represents the amount of smoke entering the smoke storage tank at time t; e WI It indicates the flue gas emission intensity per unit of waste incineration power generation system (t / kW·h); P WI (t) represents the power generation of waste incineration at time t (kW). Considering that the total power consumed by flue gas treatment is composed of the power consumed by the air pump operation and the power consumed by flue gas purification.

[0195]

[0196] Where: P WI，s (t) represents the total power consumption of flue gas treatment at time t (kW); μ WI，s Indicates the unit energy consumption coefficient of flue gas treatment; μ GP Indicates the energy consumption of the air pump when the gas enters the flue gas; It indicates the amount of smoke in the smoke storage tank entering the reaction tower at time t.

[0197] In step S102, based on the internal output unit model, an objective function and corresponding constraints are constructed with the goal of minimizing the operating cost of the urban virtual power plant, minimizing carbon emissions, maximizing demand response benefits and maximizing robustness.

[0198] Objective function: The objective function is to minimize VPP operating costs, minimize carbon emissions, and maximize DR benefits, that is, to meet the following requirements:

[0199]

[0200] Where: C(P g ) represents the power generation cost of micro gas turbine, C OP represents the sum of the operation and maintenance costs of wind power, photovoltaic electrolyzers, fuel cells, waste incineration generators and flue gas purification devices, (P b ) represents the cost of electricity purchase and sale between VPP and the power grid, C c represents the carbon trading cost, E c represents carbon emissions, U represents the benefit of demand response, and Ψ is the robustness of a scheduling cycle.

[0201] 1. VPP operating cost: When considering economic efficiency, the objective function is to minimize the VPP operating cost. The VPP operating cost mainly includes the power generation cost of the power generator, the operation and maintenance cost, and the carbon emission rights trading cost. The first objective function is:

[0202] min F1=C(P g )+C OP +C c (18)

[0203] Where: C(P g ) represents the power generation cost of micro gas turbine, C OP represents the operation and maintenance costs of wind power, photovoltaic electrolyzer, fuel cell, waste incineration generator set and flue gas purification device, C c represents the carbon trading cost.

[0204] The specific expression of each part:

[0205]

[0206] In the formula, C i (P g,i,t ) is the cost function of micro gas turbine, P g,i,t is the power generation of the i-th micro gas turbine at time t, a i , b i 、c i are the power generation cost coefficients of the i-th micro gas turbine, N is the total number of micro gas turbines, β WT , βPV , β ele , β fc , β WI , β WI,s They represent the operating cost coefficients of wind power, photovoltaic power, electrolyzer, fuel cell, waste incineration generator set, and flue gas purification device, respectively. WT,t , P PV,t , P ele,n,t , P fc,t , P WI (t), P WI,s (t) are the outputs of wind power, photovoltaic power, electrolyzer, fuel cell, waste incineration generator set and flue gas purification device at time t (kW), λ c,t is the price of unit carbon emission rights at time t, ΔE represents the difference in carbon emissions, and κ i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine, κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; γ is the VPP carbon emission quota allocation coefficient, P WT,t represents the wind power at time t; P PV,t Represents the photovoltaic power generation at time t.

[0207] 2. VPP carbon emissions: When considering low carbon, the goal is to minimize carbon emissions. The second objective function is:

[0208]

[0209] Where: P g,i,t is the power generation of the i-th micro gas turbine at time t, κ i1 is the carbon emission intensity per unit of electricity of the i-th micro gas turbine, P WI，t is the power generation of the i-th waste incineration power generation unit at time t, κ i2 is the carbon emission intensity per unit electricity of the i-th waste incineration power generation unit.

[0210] 3. DR benefit: Incentive demand response requires VPP operators to pay subsidies to responding users, that is, incentive costs, which also reflect the benefits on the user side. In addition, traditional units in VPP will gain cost benefits due to less power generation. Taking the maximum benefit of VPP demand response as the optimization goal, the third objective function is:

[0211]

[0212] Where: U is the benefit of demand response, x i,t Reduce the power consumption of the i-th DR user at time t, λ i x i,t represents the response x of the i-th DR user at time t i,tWhen the power is less, the marginal benefit that VPP obtains due to less power generation; n i,t represents the subsidy paid to the i-th DR user at time t, m 1,i 、m 2,i is the incentive cost coefficient of the i-th type DR user, θ i is the response willingness of the i-th DR user.

[0213] 4. Robustness: Virtual power plants usually contain a variety of distributed energy sources (such as wind energy and solar energy), the output of which is intermittent and volatile. Virtual power plants with high robustness can better cope with these uncertainties and ensure the stability and reliability of power supply. Taking robustness as the optimization goal, the fourth objective function is:

[0214]

[0215] Where: is the cumulative probability corresponding to the upper and lower bounds of the robust operating interval.

[0216] Constraints. Constraints include: load supply and demand balance constraints, hydrogen energy balance constraints, micro gas turbine output constraints, wind and solar output constraints, electrolyzer constraints, hydrogen storage tank constraints, fuel cell power constraints, waste incineration power generation unit power constraints, flue gas purification device power constraints and demand response constraints, among which, the load supply and demand balance constraint is that the sum of the power output power of the urban virtual power plant at time t meets all power demands at time t; the hydrogen energy balance constraint is that at each time point all the hydrogen produced is equal to all the hydrogen consumed to ensure the balance of hydrogen supply and demand; the micro gas turbine output constraint is to ensure that the micro gas turbine operates within a safe and economic range by setting the output range of each micro gas turbine and the output change rate in adjacent time intervals; the wind and solar output constraint is that the output of wind power and photovoltaic power generation is set between 0 and its maximum output to ensure that wind turbines and photovoltaic generators operate within a safe and economic range. and within the scope of economy; the electrolyzer constraint is to set the power constraint of the electrolyzer and the start-stop constraint of the electrolyzer to ensure the safe and reliable operation of the electrolyzer; the hydrogen storage tank constraint is to set the upper and lower limit constraints and continuity constraint of the hydrogen storage tank; the fuel cell power constraint is to set the fuel cell power between 0 and the maximum power of the fuel cell; the waste incineration power generation unit power constraint is to set the waste incineration power generation unit power between 0 and the maximum power of the waste incineration power generation unit; the flue gas purification device power constraint is to set the flue gas purification device power between 0 and the maximum power of the flue gas purification device; the waste resource utilization potential constraint is to reasonably arrange the operation plan of the waste incineration power generation unit according to the change of the utilization potential of the waste resources at time t during the scheduling process to ensure that its output is within the system demand range; and the demand response constraint is to set the load reduction of each type of user demand response within a reasonable range.

[0217] 1. Load supply and demand balance constraint: The sum of the power output powers of the virtual power plant at time t must meet all power demands at that time, that is, the output power of wind power, photovoltaics, micro gas turbines, fuel cells, and garbage power generation units must be able to meet the power consumption of power loads, electrolyzers, and flue gas purification devices.

[0218] P WT,t +P PV,t +P g,t +P fc,t +P WI,t =L t +P ele,i,t +P WI,s,t (25)

[0219] Where: P WT,t represents the wind power at time t; P PV,t represents the photovoltaic power generation at time t, P g,t represents the power of the micro gas turbine at time t, P fc,t represents the power of the fuel cell at time t, P WI,t represents the power of the waste incineration generator set at time t; L t Indicates the power load, P ele,i,t represents the power of the electrolytic cell at time t, P WI,s,t Represents the power of the flue gas purification device at time t.

[0220] 2. Hydrogen energy balance constraint: To ensure the balance between supply and demand of hydrogen, that is, at each point in time, all hydrogen produced (including the lost part) must be equal to all hydrogen consumed (including storage in hydrogen storage tanks and use of fuel cells).

[0221] M ele,t +M loss,t =L h,t +M sto,t +M fc,t (26)

[0222]

[0223] Where: M ele,t represents the mass of hydrogen produced by the electrolyzer at time t; M loss,t Indicates the mass of hydrogen lost at time t; L h,t represents the hydrogen load at time t; M sto,t M represents the mass of hydrogen absorbed or released by the hydrogen storage tank at time t, where absorption is positive and release is negative; fc,t represents the mass of hydrogen consumed by the fuel cell at time t; η H2E Electrolyzer efficiency; HHV represents the heating value of hydrogen, which is 39.44 kW·h / kg.

[0224] 3. Micro gas turbine output constraint: To ensure that the micro gas turbine operates within a safe and economical range, it is necessary to set the output range of each micro gas turbine and the output change rate in adjacent time intervals. The minimum output limit prevents the equipment from running at no load, the maximum output limit avoids overload operation, and the output change rate ensures the smooth operation of the equipment and prevents mechanical damage or system instability caused by rapid increase or decrease in load.

[0225]

[0226] P g,i,min ≤P g,i,t ≤P g,i,max (28)

[0227] P d,i ≤P g,i,t+1 -P g,i,t ≤P u,i (32)

[0228] Where: P g,i,min , P g,i,max Respectively represent the lower and upper limits of the output of the i-th micro gas turbine; P d,i , P u,i They represent the downhill climbing and uphill climbing of the i-th micro gas turbine respectively.

[0229] 4. Wind and solar output constraints: The output of wind power and photovoltaic power generation must be between 0 and its maximum output to ensure that they operate within a safe and economical range.

[0230] 0≤P WT,t ≤P WT,max (33)

[0231] 0≤P PV,t ≤P PV,max (34)

[0232] Where: P WT,max , P PV,max They are the maximum outputs of wind power and photovoltaic power respectively.

[0233] 5. Electrolyzer constraints: In order to ensure the safe and reliable operation of the electrolyzer and give full play to the performance of the electrolyzer, the electrolyzer constraints include the power constraints of the electrolyzer and the start and stop constraints of the electrolyzer.

[0234] The electrolyzer power constraint is

[0235] P ele,min U ele,i,t ≤P ele,i,t ≤P ele,max U ele,i,t (29)

[0236] Where: Pele,min and P ele,max are the minimum and maximum power of the electrolyzer, in kW. In this paper, the minimum and maximum power of the electrolyzer are set to 25% and 100% of the rated power, respectively; U ele,i,t It is the working status flag of the i-th electrolytic cell at time t, 1 means running and 0 means stopping.

[0237] The start and stop constraints of the electrolyzer are

[0238]

[0239] Where: Y i,t , Z i,t Indicates the start and stop status of the i-th electrolytic cell at time t. ON indicates the maximum number of starts allowed per day for the electrolytic cell, and OFF indicates the maximum number of stops allowed per day for the electrolytic cell.

[0240] 7. Hydrogen storage tank constraints: Hydrogen storage tank constraints include upper and lower limit constraints and continuity constraints. Upper and lower limit constraints of hydrogen storage tanks:

[0241] SOHC min ≤SOHC t ≤SOHC max (38)

[0242] Where: SOHC min and SOHC max They are the minimum and maximum thresholds of the hydrogen storage state of the hydrogen storage tank; SOHC t is the hydrogen storage state of the hydrogen storage tank at time t.

[0243] Hydrogen storage tank continuity constraints

[0244]

[0245] Where: H max Represents the total capacity of the hydrogen storage tank, in kg; SOHC init and SOHC final Respectively represent the SOHC value of the hydrogen storage tank at the initial and final moments of the day.

[0246] 8. Fuel cell power upper and lower limits:

[0247] 0≤P fc,t ≤P fc,max (40)

[0248] Where: P fc,max Indicates the maximum power of the fuel cell.

[0249] 9. Power constraints of waste incineration power generation units:

[0250] 0≤P WI,t ≤P WI,max (41)

[0251] Where: P WI,max Indicates the maximum power of the waste incineration generator set.

[0252] 10. Flue gas purification device:

[0253] 0≤P WI,s,t ≤P WI,s,max (30)

[0254] Where: P WI,s,max Indicates the maximum power of the flue gas purification device.

[0255] 11. Constraints on the available potential of garbage resources: In the power system dispatch, it is necessary to reasonably arrange the operation plan of the garbage incineration power generation unit according to the changes in the available potential of garbage resources at time t to ensure that its output is within the system demand range.

[0256]

[0257] Where: They represent the minimum and maximum supply of garbage resources in the cycle (kg); They are the minimum combustible waste ratio and the maximum combustible waste ratio.

[0258] 12. Demand response constraints: Demand response requires that the load reduction of each type of user is within a reasonable range. The minimum reduction limit prevents users from unnecessary load reduction, and the maximum reduction limit avoids the inconvenience caused by excessive load reduction. The total load reduction of each type of user in the entire scheduling cycle shall not exceed the maximum daily load reduction of the demand response of this type of user, ensuring that users will not affect their normal life or production activities due to excessive participation in demand response.

[0259]

[0260] x i,min ≤x i,t ≤x i,max (45)

[0261]

[0262] Where: D t represents the sum of the load reductions of all user types in response to demand at time t, x i,t represents the load reduction of the i-th user at time t, x i,min 、x i,maxThey represent the minimum reduction and maximum reduction at any time point t, M represents the total number of user types; D L,i It represents the maximum total load reduction of the i-th user in response to demand reduction. It is the value range of user type i.

[0263] The net benefit of users participating in demand response is non-negative, that is, the subsidy obtained by users can at least cover the outage costs caused by demand response. At the same time, to ensure fairness among different user types, higher priority user types (for example, important users or sensitive users) should receive higher net benefits to encourage them to participate in demand response activities first.

[0264]

[0265]

[0266] Where: n i,t is the subsidy paid by VPP to the i-th user at time t, It represents the outage cost of the i-th type of demand response user.

[0267] In step S103, a chance-constrained distributed robustness model is constructed based on the wind and solar power output prediction error fuzzy set.

[0268] In order to achieve flexible and adjustable robustness of the scheduling scheme and fully consider the impact of the cumulative probability distribution of wind and solar power forecast errors and their deviations on the scheduling results, the wind and solar power output forecast error fuzzy set and chance-constrained distributed robustness model are constructed based on the imprecise Dirichlet model (IDM) theory.

[0269] 1. Construct the prediction error fuzzy set based on the IDM method. Under the confidence level γ, the true probability θ interval is constructed as follows:

[0270]

[0271] Where H and G are the Beta distribution B(m i ,s+nm i ) and Beta distribution B(s+m i ,nm i )’s cumulative distribution function; θ1 and θ′ are the upper and lower bounds of the true probability distribution of θ; m i and n represent the number of samples where the random variable ξ is less than sample A and the total number of samples, respectively. The fuzzy set can be expressed as:

[0272] P={F∈P0(|ξ1,ξ′|)|F[ξ<ξ i ]∈[θ 1i<θ i ′],i=1,2,…,n} (50)

[0273] Where: P0(|ξ1,ξ′|) represents the set of all probabilities within the estimation range |ξ1,ξ′|; F is the sample point ξ i The cumulative probability distribution of .

[0274] 2. Opportunity constraint distribution robustness: Based on the establishment of the fuzzy set of wind and solar forecast errors, the system is further constructed to provide upper and lower backup The robust operation range and opportunity-constrained distribution robustness model are proposed to describe the tolerance range and robustness of the system to the disturbance caused by the wind and solar forecast errors. The opportunity-constrained distribution robustness model is as follows.

[0275]

[0276] Where: Ψ is the robustness of a scheduling cycle; Γ t is the robustness corresponding to the robust operation interval at time t; R is the robust interval set; R t for robust operation intervals and interval variables; are the upper and lower bounds of the interval respectively; P r Indicates that the time point t falls into the robust interval The probability of is the cumulative probability corresponding to the upper and lower bounds of the robust operation interval; F WP,t (·) is the cumulative probability distribution function of wind power forecast error at time t; Provide upper and lower spare capacity for coal-fired units at time t; K w Wind power installed capacity.

[0277] Since the wind and solar forecast errors occur simultaneously, it is necessary to construct the joint probability distribution of wind and solar forecast errors. Furthermore, the chance-constrained distributed robustness model under the joint probability distribution of wind and solar forecast errors is constructed as follows.

[0278]

[0279] In the formula, F 1,t (·), F′ t (·) is the upper and lower bounds of the joint cumulative probability distribution of the wind-solar joint prediction error at time t.

[0280] In step S104, the multi-objective objective function is converted into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model.

[0281] 3. Multi-objective optimal compromise economic low-carbon benefit optimization model: Combining multi-objective decision-making theory and TOPSIS method, by scientifically and objectively normalizing the optimal compromise processing of economic, low-carbon and benefit indicators, constructing the economic low-carbon benefit multi-objective compromise function, and transforming the multi-objective optimization problem into the optimal compromise single-objective optimization problem. The construction of the multi-objective optimal compromise function mainly has the following three steps.

[0282] (1) First, to deal with the inconsistency of the objective function’s economic efficiency, low carbon efficiency, and benefit dimensions, equations (18), (23), (24), and (26) are normalized as follows.

[0283]

[0284] In the formula, F represents the vector form of system economy, low carbon, demand response benefit, and robustness; F i represents the i-th target; are the minimum and maximum values ​​of the i-th target, respectively, Z i Represents the normalized value of the i-th target.

[0285] (2) According to the above formula, we can obtain the normalized optimal solution (0, 0, 1, 1) and the worst solution (1, 1, 0, 0). We can further calculate the relative distance between the scheduling scheme Z = [Z1, Z2, Z3, Z4] and the optimal solution and the worst solution.

[0286]

[0287] Where: D + , D - They respectively represent the relative distances of the system scheduling solution from the optimal solution and the worst solution.

[0288] (3) Construct a multi-objective optimal compromise function. The scheduling results obtained by the system can be judged by the closer the distance to the optimal solution and the farther the distance to the worst solution. The multi-objective optimal compromise function is constructed according to the formula (2) as follows.

[0289]

[0290] Where: W is the compromise optimization degree, 0≤W≤1, the closer W is to 1, the better the solution.

[0291] In step S105, a multi-objective optimal compromise economic low-carbon benefit optimization model is used to obtain an optimal compromise solution, a corresponding confidence level and a scheduling plan.

[0292] 4. Solution of multi-objective optimal compromise robust scheduling model

[0293] In view of the nonlinear and high-dimensional characteristics of the constructed multi-objective optimal compromise scheduling model, this paper adopts the adaptive hybrid composite differential evolution algorithm (AHA-CDE) to solve it. The solution process Figure 5 As follows: (1) Input the parameters of each device, historical data of wind and solar power, and forecast data of wind and solar power, set the algorithm parameters, and initialize the population (composed of randomly generated individuals of each system scheduling scheme); (2) Generate confidence bands under the confidence level γ for historical data of wind and solar power based on the IDM method; (3) Calculate the operating cost F1, carbon emissions F2, demand response benefit F3 and robustness F4, and use the TOPSIS method to calculate the compromise optimization degree W value of each target; (4) Evolved individuals are sorted in descending order according to the compromise optimization degree W value; (5) Use the AHA-CDE algorithm to obtain a new population of offspring; (6) Repeat steps (3)-(5) and iterate continuously until Gmax is satisfied; (7) Obtain the optimal compromise optimization degree W value, the optimal compromise solution and the optimal compromise scheduling scheme of the system.

[0294] The technical solution of the present invention is based on a hydrogen energy storage urban virtual power plant. By introducing a waste incineration unit and proposing a multi-objective decision-making theory and approximate ideal solution sorting technology, a multi-objective optimal compromise function that comprehensively considers the economy, low carbon, demand response benefits and robustness of system scheduling is constructed, thereby establishing a multi-objective optimal compromise distributed robust scheduling model containing hydrogen energy storage. The optimized scheduling plan of the hydrogen energy storage urban virtual power plant with economy, low carbon, efficiency and robustness can be quickly and accurately obtained, which can better guide the operation and trading of the hydrogen energy storage virtual power plant, improve carbon emission reduction and energy utilization efficiency, and thus promote the construction of my country's new power system.

[0295] refer to Figure 6 On the other hand, a distributed robust optimization scheduling system for urban virtual power plants is provided, characterized in that it includes: an output unit model construction module 601, which is used to construct an internal output unit model of the urban virtual power plant, wherein the internal output unit model includes an electrolysis hydrogen production model, a hydrogen storage system model, a fuel cell model, a waste incineration power generation unit output model and a flue gas purification device output model; an objective function construction module 602, which is used to construct an objective function and corresponding constraints based on the internal output unit model with the objectives of minimizing the operating cost, minimizing the carbon emissions, maximizing the demand response benefit and maximizing the robustness of the urban virtual power plant; a robustness model construction module 603, which is used to construct an opportunity-constrained distributed robustness model based on the fuzzy set of wind and solar power output prediction errors; a conversion module 604, which is used to convert the multi-objective objective function into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model; and a solution module 605, which is used to use the multi-objective optimal compromise economic low-carbon benefit optimization model to obtain the optimal compromise solution, the corresponding confidence level and the scheduling plan.

[0296] The objective function construction module 602 is also used to construct a first objective function, a second objective function, a third objective function and a fourth objective function, wherein the first objective function of minimizing the operating cost of the urban virtual power plant is represented by formulas (18)-(22); the second objective function of minimizing the carbon emissions of the urban virtual power plant is represented by formula (23); the third objective function of maximizing the demand response benefit of the urban virtual power plant is represented by the following formulas (24) and (25); and the fourth objective function of maximizing the robustness is represented by the following formula (26).

[0297] The conversion module 604 includes a normalization processing module, a relative distance determination module and a multi-objective optimal compromise function construction module. The normalization processing module is used to normalize the first objective function, the second objective function, the third objective function and the fourth objective function through formula (53). The relative distance determination module is used to obtain the normalized optimal solution and the worst solution according to the above normalization processing formula, so as to calculate the relative distance of the scheduling scheme Z from the optimal solution and the worst solution through formula (54). The multi-objective optimal compromise function construction module is used to construct a multi-objective optimal compromise function based on the relative distance of the system scheduling scheme from the optimal solution and the worst solution (refer to formula (55)).

[0298] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0299] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for optimizing the scheduling of urban virtual power plants, characterized in that: include: Constructing an internal output unit model of a town virtual power plant, wherein the internal output unit model includes an electrolysis hydrogen production model, a hydrogen storage system model, a fuel cell model, a waste incineration power generation unit output model, and a flue gas purification device output model; Based on the internal output unit model, an objective function and corresponding constraints are constructed with the goal of minimizing the operating cost, minimizing carbon emissions, maximizing demand response benefits and maximizing robustness of the urban virtual power plant; Based on the fuzzy set of wind and solar power output forecast error, a chance-constrained distributed robustness model is constructed. Converting the multi-objective objective function into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model; and The multi-objective optimal compromise economic low-carbon benefit optimization model is used to obtain the optimal compromise solution, the corresponding confidence level and the scheduling plan.

2. The method for optimizing the scheduling of urban virtual power plants according to claim 1 is characterized in that: The objective function constructed based on the internal output unit model with the goal of minimizing the operating cost of the urban virtual power plant, minimizing the carbon emissions, maximizing the demand response benefit and maximizing the robustness further includes: The first objective function of minimizing the operating cost of the town virtual power plant is expressed by the following formula: minF1=C(P g )+C OP +C c ; Among them, C(P g ) represents the power generation cost of the micro gas turbine; C OP represents the operation and maintenance costs of wind power, photovoltaic electrolyzers, fuel cells, waste incineration generators and flue gas purification devices; C c represents the carbon trading cost; C i (P g,i,t ) is the cost function of micro gas turbine, P g,i,t is the power generation of the i-th micro gas turbine at time t; a i 、b i 、c i are the power generation cost coefficients of the i-th micro gas turbine, N is the total number of micro gas turbines; β WT , β PV , β ele , β fc , β WI , β WI,s Respectively represent the operating cost coefficients of wind power, photovoltaic, electrolyzer, fuel cell, waste incineration generator set, and flue gas purification device; P WT,t , P PV,t , P el e ,n,t , P fc,t , P WI (t), P WI,s (t) are the output of wind power, photovoltaic power, electrolyzer, fuel cell, waste incineration generator set and flue gas purification device at time t obtained based on the models of each internal output unit; λ c,t is the price of unit carbon emission rights at time t; ΔE represents the difference in carbon emissions; κ i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; γ is the carbon emission quota allocation coefficient of the urban virtual power plant; P WT,t represents the wind power at time t; P PV,t represents the photovoltaic power generation at time t; The second objective function of the town virtual power plant with the lowest carbon emissions is expressed by the following formula: Among them, P g,i,t is the power generation of the i-th micro gas turbine at time t; κ i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; P WI,t is the power generation of the i-th waste incineration power generation unit at time t; κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; The third objective function that maximizes the demand response benefit of the urban virtual power plant is expressed by the following formula: Where U is the benefit of demand response, x i,t The power reduction of the i-th type of demand response user at time t, λ i x i,t represents the user response x of the i-th type of demand response at time t i,t When the power consumption is less, the marginal benefit of the urban virtual power plant due to less power generation; n i,t represents the subsidy paid to the i-th type of demand response user at time t, m 1,i 、m 2,i is the incentive cost coefficient of the i-th type of demand response user, θ i The response willingness of the i-th type of demand response user; and The fourth objective function with the highest robustness is expressed by the following formula: in, is the cumulative probability corresponding to the upper and lower bounds of the robust operating interval.

3. The method for optimizing the scheduling of urban virtual power plants according to claim 2 is characterized in that: The multi-objective objective function is converted into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model further including: The first objective function, the second objective function, the third objective function and the fourth objective function are normalized by the following formula: Among them, F represents the vector form of system economy, low carbon and demand response benefits; F i represents the i-th target; are the minimum and maximum values ​​of the i-th target, respectively, Z i Represents the normalized value of the i-th target; According to the above normalization processing formula, the normalized optimal solution and the worst solution are obtained, and the relative distance of the scheduling scheme Z from the optimal solution and the worst solution is calculated by the following formula: Among them, D + , D - Respectively represent the relative distances of the system scheduling solution from the optimal solution and the worst solution; Based on the relative distances of the system scheduling solution from the optimal solution and the worst solution, the following multi-objective optimal compromise function is constructed: Wherein, W is the compromise optimization degree, 0≤W≤1, and the merits of the scheme are judged according to the compromise optimization degree.

4. The method for optimizing the scheduling of urban virtual power plants according to claim 3 is characterized in that: The constraints include: load supply and demand balance constraints, hydrogen energy balance constraints, micro gas turbine output constraints, wind and solar output constraints, electrolyzer constraints, hydrogen storage tank constraints, fuel cell power constraints, waste incineration power generation unit power constraints, flue gas purification device power constraints and demand response constraints, among which, The load supply and demand balance constraint is that the sum of the power output powers of the urban virtual power plant at time t satisfies all power demands at time t; The hydrogen energy balance constraint is that at each time point, all hydrogen produced is equal to all hydrogen consumed to ensure the balance between hydrogen supply and demand; The micro gas turbine output constraint is to ensure that the micro gas turbine operates within a safe and economical range by setting the output range of each micro gas turbine and the output change rate in adjacent time intervals; The wind and solar power output constraint is that the output of wind power and photovoltaic power generation is set between 0 and its maximum output to ensure that the wind turbine and photovoltaic power generation set operate within a safe and economical range; The electrolytic cell constraint is to set the power constraint of the electrolytic cell and the start and stop constraint of the electrolytic cell to ensure the safe and reliable operation of the electrolytic cell; The hydrogen storage tank constraint is to set the upper and lower limit constraints and continuity constraints of the hydrogen storage tank; The fuel cell power constraint is to set the fuel cell power between 0 and the maximum power of the fuel cell; The power constraint of the waste incineration generator set is to set the power of the waste incineration generator set between 0 and the maximum power of the waste incineration generator set; The power constraint of the flue gas purification device is to set the power of the flue gas purification device between 0 and the maximum power of the flue gas purification device; The waste resource utilization potential constraint is to reasonably arrange the operation plan of the waste incineration power generation unit according to the change of the utilization potential of the waste resources at time t during the scheduling process to ensure that its output is within the system demand range; and The demand response constraint is to set the load reduction of each type of user demand response within a reasonable range.

5. The method for optimizing the scheduling of urban virtual power plants according to claim 4 is characterized in that: The load supply and demand balance constraint is calculated by the following formula: P WT,t +P PV,t +P g,t +P fc,t +P WI,t =L t +P ele,i,t +P WI,s,t ; Among them, P WT,t represents the wind power at time t; P PV,t represents the photovoltaic power generation at time t, P g,t represents the power of the micro gas turbine at time t, P fc,t represents the power of the fuel cell at time t, P WI,t Indicates the power of the waste incineration generator set at time t; L t Indicates the power load, P ele,i,t represents the power of the electrolytic cell at time t, P WI,s,t Indicates the power of the flue gas purification device at time t; The hydrogen energy balance constraint is calculated by the following formula: M ele,t +M loss,t =L h,t +M sto,t +M fc,t Among them, M ele,t represents the mass of hydrogen produced by the electrolyzer at time t; M loss,t represents the mass of hydrogen lost at time t; L h,t represents the hydrogen load at time t; M sto,t Represents the mass of hydrogen absorbed or released by the hydrogen storage tank at time t, with absorption being positive and release being negative; M fc,t represents the mass of hydrogen consumed by the fuel cell at time t; η H2E Electrolyzer efficiency; HHV represents the heating value of hydrogen, which is 39.44 kW·h / kg; The micro gas turbine output constraint is calculated by the following formula: P g,i，min ≤P g,i,t ≤P g,i,max ; P d，i ≤P g,i,t+1 -P g，i，t ≤P u，i ; Among them, P g,t represents the power of the micro gas turbine at time t; P g,i,t represents the power of the i-th micro gas turbine at time t; P g,i,min , P g,i,max Respectively represent the lower and upper limits of the output of the i-th micro gas turbine; P d,i , P u,i They represent down-slope and up-slope climbing of the i-th micro gas turbine respectively; The wind and solar output constraint is calculated by the following formula: 0≤P WT,t ≤P WT,max ; 0≤P PV,t ≤P PV,max ; Among them, P WT,max , P PV,max They are the maximum outputs of wind power and photovoltaic power respectively; The power constraint and the start-stop constraint of the electrolyzer are calculated by the following formula: P ele,min IN ele,i,t ≤P ele,i,t ≤P ele,max IN ele,i,t ; Among them, P ele,min and P ele,max are the minimum and maximum power of the electrolyzer, in kW, and the minimum and maximum power of the electrolyzer are set to 25% and 100% of the rated power respectively; U ele,i,t is the working status flag of the i-th electrolytic cell at time t, 1 indicates operation, 0 indicates shutdown; Y i,t , Z i,t Indicates the start and stop status of the i-th electrolytic cell at time t; ON indicates the maximum number of starts allowed per day for the electrolytic cell, and OFF indicates the maximum number of stops allowed per day for the electrolytic cell; The upper and lower limit constraints and continuity constraints of the hydrogen storage tank are calculated by the following formula: SOHC min ≤SOHC t ≤SOHC max ; Among them, SOHC min and SOHC max They are the minimum and maximum thresholds of the hydrogen storage state of the hydrogen storage tank; SOHC t is the hydrogen storage state of the hydrogen storage tank at time t; H max Represents the total capacity of the hydrogen storage tank, in kg; SOHC init and SOHC final Respectively represent the SOHC value of the hydrogen storage tank at the initial and final moments of a day; The upper and lower limits of the fuel cell power are calculated by the following formula: 0≤P fc,t ≤P fc,max ; Among them, P fc,max Indicates the maximum power of the fuel cell; The power constraint of the waste incineration power generation unit is calculated by the following formula: 0≤P WI,t ≤P WI,max ; Among them, P WI,max Indicates the maximum power of the waste incineration generator set; The power constraint of the flue gas purification device is calculated by the following formula: 0≤P WI,s，t ≤P WI,s,max Among them, P WI,s,max Indicates the maximum power of the flue gas purification device; The waste resource utilization potential constraint is calculated by the following formula: in, They represent the minimum and maximum supply of garbage resources in the cycle respectively; They are the minimum combustible waste ratio and the maximum combustible waste ratio; The demand response constraint is calculated by the following formula: x i，min ≤x i，t ≤x i，max ; Among them, D t represents the sum of load reductions in response to demand by all user types at time t; x i,t represents the load reduction of the i-th user at time t; x i,min 、x i,max They represent the minimum reduction and maximum reduction at any time point t; M represents the total number of user types; D L,i It represents the maximum total load reduction of the i-th user in response to demand reduction on a daily basis; The value range of user type i; Among them, n i,t is the subsidy paid by the urban virtual power plant to the i-th user at time t, It represents the outage cost of the i-th type of demand response user.

6. The method for optimizing the scheduling of urban virtual power plants according to claim 5 is characterized in that: The internal output unit model also includes a gas turbine model, a wind power output model and a photovoltaic output model, wherein: The gas turbine model is expressed by the following formula: P g,min ≤P g ≤P g,max ; P d ≤P g,t+1 -P g,t ≤P u ; Among them, C g represents the power generation cost of micro gas turbine, a, b, c are power generation cost coefficients, P g is the power generation power of the micro gas turbine; g,min , P g,max Respectively represent the lower and upper limits of gas turbine output, P g,t , P g,t+1 Respectively represent the output power at time point t and the next time point t+1, P d , P u They represent the maximum down-ramp rate and up-ramp rate acceptable to the gas turbine respectively; The wind power output model is expressed by the following formula: Among them, v ci is the cut-in wind speed; v co is the cut-out wind speed; v r is the rated wind speed; v is the natural wind speed, P WT Indicates the active output power of the fan; P WT,r Indicates the rated active power of the fan; The photovoltaic output model is expressed by the following formula: Among them, R c Indicates the actual solar radiation intensity, R max Indicates the maximum solar radiation intensity, P PV Represents the photovoltaic active output power, P max , T max At 1000W / m 2 , maximum output power and maximum reference temperature under 25℃ standard conditions, T c represents the actual operating temperature, and k represents the power temperature coefficient; The electrolysis hydrogen production model is expressed by the following formula: M ele =0.017P ele +0.22; Among them, M ele Indicates the mass of hydrogen produced per hour, in kg; P ele is the electrolytic cell power; The hydrogen storage state SOHC of the hydrogen storage tank in the hydrogen storage system model is expressed by the following formula: M sto,t =M ele,t -M fe,t -L h,t ; Among them, M sto,t Indicates the mass of hydrogen filled or discharged in the hydrogen storage tank at time t, in kg, M sto,t A positive value indicates that the hydrogen storage tank is filled with hydrogen at time t, M sto,t Negative indicates that the hydrogen storage tank releases hydrogen at time t, H max It represents the capacity of the hydrogen storage tank in kg; Δt represents a time interval; M ele,t represents the mass of hydrogen produced by the electrolyzer at time t, L h,t represents the hydrogen load at time t, M fe,t Indicates the mass of hydrogen consumed by the fuel cell power generation at time t, which is always 0 in the grid-connected scenario; The power generation of the proton exchange membrane fuel cell PEMFC in the fuel cell model is expressed by the following formula: Among them, P fc (t) represents the power generation of the fuel cell at time t, V H (t) represents the hydrogen consumption of the fuel cell at time t, Q H Indicates the calorific value of hydrogen, η fc-e Indicates the efficiency of fuel cell power generation; The output model of the waste incineration power generation unit is expressed by the following formula: P WI (t)=λ WI ·R WI (t); P WI,n (t)=P WI (t)-P WI，s (t); Among them, P WI (t) represents the power generation of waste incineration at time t; WI Indicates the energy conversion coefficient of combustible waste. The electricity converted per ton of domestic waste entering the factory is 280Wh / t; R WI (t) represents the amount of combustible waste burned (kg); P WI，n (t) is the net electricity generated by the waste incineration power generation system; P WI，s (t) represents the total power consumption of flue gas at time t (kW); The output model of the flue gas purification device is expressed by the following formula: F sk (t)=P WI (t)·e WI =F sk,cl (t)+F sk,s (t); Where, ρ represents the smoke split ratio (%); F sk (t) represents the total flue gas volume generated by waste incineration power generation at time t; F sk,cl (t) represents the amount of flue gas directly treated at time t; F sk,s (t) represents the amount of smoke entering the smoke storage tank at time t; e WI It represents the flue gas emission intensity per unit of waste incineration power generation system (t / kW·h); P WI (t) represents the power generation of waste incineration at time t kW; P WI,s (t) represents the total power consumption of flue gas treatment at time t kW; μ WI,s Indicates the unit energy consumption coefficient of flue gas treatment; μ GP Indicates the energy consumption of the air pump when the gas enters the flue gas; It indicates the amount of smoke in the smoke storage tank entering the reaction tower at time t.

7. The method for optimizing the scheduling of urban virtual power plants according to claim 4 is characterized in that: The opportunity constraint distribution robustness model based on the fuzzy set of wind and solar power output forecast error further includes: At the confidence level γ, the true probability θ interval is as follows: Among them, H and G are the Beta distribution B(m i ,s+nm i ) and Beta distribution B(s+m i ,nm i )’s cumulative distribution function; θ1 and θ′ are the upper and lower bounds of the true probability distribution of θ; m i and n represent the number of samples and the total number of samples where the random variable ξ is less than sample A respectively; the fuzzy set of wind and solar power output prediction error is expressed as: P={F∈P0(|ξ1,ξ′|)|F[ξ<ξ i ]∈[θ 1i <θ′ i ],i=1,2,...,n}; Among them, P0(|ξ1,ξ′|) represents the set of all probabilities within the estimation range |ξ1,ξ′|; F is the sample point ξ i The cumulative probability distribution of Based on the fuzzy set of wind and solar power output prediction errors, the following opportunity-constrained distributed robustness model is constructed: Among them, Ψ is the robustness of a scheduling cycle; Γ t is the robustness corresponding to the robust operation interval at time t; R is the robust interval set; R t are the robust operating interval and interval variable respectively; are the upper and lower bounds of the interval respectively; P r Indicates that the time point t falls into the robust interval The probability of is the cumulative probability corresponding to the upper and lower bounds of the robust operation interval; F WP,t (·) is the cumulative probability distribution function of wind power forecast error at time t; Provide upper and lower spare capacity for coal-fired units at time t; K w for wind power installed capacity; The chance-constrained distributed robustness model under the joint probability distribution of wind and solar forecast errors is constructed as follows: Among them, F 1,t (·), F′ t (·) is the upper and lower bounds of the joint cumulative probability distribution of the wind-solar joint prediction error at time t.

8. A distributed optimization dispatching system for urban virtual power plants, characterized in that: include: An output unit model construction module is used to construct an internal output unit model of a town virtual power plant, wherein the internal output unit model includes an electrolysis hydrogen production model, a hydrogen storage system model, a fuel cell model, a waste incineration power generation unit output model, and a flue gas purification device output model; An objective function construction module is used to construct an objective function and corresponding constraints based on the internal output unit model with the goal of minimizing the operating cost, minimizing carbon emissions, maximizing demand response benefits and maximizing robustness of the urban virtual power plant; A robustness model building module, which is used to build a chance-constrained distributed robustness model based on the fuzzy set of wind and solar output forecast errors; and A conversion module, used for converting the multi-objective objective function into an optimal compromise single objective function through normalized optimal compromise processing to obtain a multi-objective optimal compromise economic low-carbon benefit optimization model; The solution module is used to obtain the optimal compromise solution, the corresponding confidence level and the scheduling plan by using the multi-objective optimal compromise economic low-carbon benefit optimization model.

9. The urban virtual power plant distributed optimization scheduling system according to claim 8 is characterized in that: The objective function construction module is also used to construct a first objective function, a second objective function, a third objective function and a fourth objective function, wherein: The first objective function of minimizing the operating cost of the town virtual power plant is expressed by the following formula: minF1=C(P g )+C OP +C c ; Among them, C(P g ) represents the power generation cost of the micro gas turbine; C OP represents the operation and maintenance costs of wind power, photovoltaic electrolyzers, fuel cells, waste incineration generators and flue gas purification devices; C c represents the carbon trading cost; C i (P g,i,t ) is the cost function of micro gas turbine, P g,i,t is the power generation of the i-th micro gas turbine at time t; a i 、b i 、c i are the power generation cost coefficients of the i-th micro gas turbine, N is the total number of micro gas turbines; β WT , β PV , β ele , β fc , β WI , β WI,s Respectively represent the operating cost coefficients of wind power, photovoltaic, electrolyzer, fuel cell, waste incineration generator set, and flue gas purification device; P WT,t , P PV,t , P ele,n,t , P fc,t , P WI (t), P WI,s (t) are the output of wind power, photovoltaic power, electrolyzer, fuel cell, waste incineration generator set and flue gas purification device at time t; λ c,t is the price of unit carbon emission rights at time t; ΔE represents the difference in carbon emissions; κ i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; γ is the carbon emission quota allocation coefficient of the urban virtual power plant; P WT,t represents the wind power at time t; P PV,t represents the photovoltaic power generation at time t; The second objective function of the town virtual power plant with the lowest carbon emissions is expressed by the following formula: Among them, P g,i,t is the power generation of the i-th micro gas turbine at time t; κ i1 is the carbon emission intensity per unit electricity of the i-th micro gas turbine; P WI,t is the power generation of the i-th waste incineration power generation unit at time t; κ i2 is the carbon emission intensity per unit of electricity of the i-th waste incineration power generation unit; The third objective function that maximizes the demand response benefit of the urban virtual power plant is expressed by the following formula: Where U is the benefit of demand response, x i,t The power reduction of the i-th type of demand response user at time t, λ i x i,t represents the user response x of the i-th demand response at time t i,t When the power consumption is less, the marginal benefit of the urban virtual power plant due to less power generation; n i,t represents the subsidy paid to the i-th type of demand response user at time t, m 1,i 、m 2,i is the incentive cost coefficient of the i-th type of demand response user, θ i The response willingness of the i-th type of demand response user; The fourth objective function with the highest robustness is expressed by the following formula: in, is the cumulative probability corresponding to the upper and lower bounds of the robust operating interval.

10. The urban virtual power plant distributed optimization dispatching system according to claim 9 is characterized in that: The conversion module is also used for: a normalization processing module, a relative distance determination module and a multi-objective optimal compromise function construction module, wherein: The normalization processing module is used to perform normalization processing on the first objective function, the second objective function, the third objective function and the fourth objective function by using the following formula: Among them, F represents the vector form of system economy, low carbon and demand response benefits; F i represents the i-th target; F i min 、F i max are the minimum and maximum values ​​of the i-th target, respectively, Z i Represents the normalized value of the i-th target; The relative distance determination module is used to obtain the normalized optimal solution and the worst solution according to the above normalization processing formula, so as to calculate the relative distance between the optimal solution and the worst solution of the scheduling scheme Z by the following formula: Among them, D + , D - Respectively represent the relative distances of the system scheduling solution from the optimal solution and the worst solution; The multi-objective optimal compromise function construction module is used to construct the following multi-objective optimal compromise function based on the relative distance between the system scheduling solution and the optimal solution and the worst solution: Wherein, W is the compromise optimization degree, 0≤W≤1, and the merits of the scheme are judged according to the compromise optimization degree.

Images