A new energy electricity consumption prediction method considering hydrogen load regulation cost

Abstract

The present invention relates to a new energy power consumption prediction method considering the hydrogen load regulation cost, and belongs to the field of new energy power generation. The method of the present invention comprises: based on the output consumption and output prediction grid-connected capacity of each photovoltaic and wind power unit, a first objective function is constructed with the maximum new energy consumption rate as the goal, and the new energy includes photovoltaic and wind power; based on the transferable hydrogen load, the shiftable hydrogen load and the interruptible hydrogen load, a second objective function is constructed with the minimum hydrogen load regulation cost as the goal; based on the hydrogen load planning curve, power balance constraints, new energy output constraints, energy storage constraints, hydrogen load constraints and other constraints, the first objective function and the second objective function are jointly solved to obtain the optimal new energy output consumption curve. The method of the present invention predicts the new energy consumption power under the background of considering the hydrogen load regulation cost, and the amount of data used in the prediction is small and the prediction accuracy is high.

Description

Technical Field

[0001] The present invention belongs to the field of renewable energy power generation, and in particular relates to a new energy electricity consumption prediction method taking into account hydrogen load regulation costs. Background Art

[0002] During the period of rapid development of new power systems, a large number of new energy power generation has been connected to the grid. Its intermittent and volatile characteristics have led to a severe situation for the consumption of new energy. In order to reduce the occurrence of wind and solar power abandonment and increase the consumption of new energy grid-connected electricity, it is necessary to coordinate with multiple parties and take multiple measures. Among them, the forecast of new energy consumption can assess the future consumption level of new energy in advance, provide an important basis for power system planning, dispatching and management, and help ensure the stable operation of the power grid, reduce power generation costs, and reduce the wind and solar abandonment rates.

[0003] Most of the existing prediction methods for renewable energy consumption are based on historical data, using historical data as training samples and machine learning to predict renewable energy consumption. This method has a certain dependence on historical data, the amount of data used for prediction is huge, and it is difficult to fully consider the future development trends of various influencing factors. When the needs of user segments change, it will greatly affect the accuracy of prediction. Summary of the invention

[0004] In view of the above analysis, the present invention aims to provide a new energy electricity consumption prediction method taking into account the hydrogen load regulation cost, and to accurately predict the grid-connected new energy consumption under the condition of taking into account the system hydrogen load regulation cost.

[0005] The method of the present invention specifically comprises the following steps:

[0006] Based on the output consumption and predicted grid-connected output of each photovoltaic and wind power unit in each period, a first objective function is constructed with the goal of maximizing the new energy consumption rate, wherein the new energy includes photovoltaic and wind power;

[0007] Based on the transferable hydrogen load, the shiftable hydrogen load and the interruptible hydrogen load in each period, a second objective function is constructed with the goal of minimizing the hydrogen load regulation cost;

[0008] Based on the hydrogen load planning curve, electricity balance constraints, new energy output constraints, energy storage constraints, hydrogen production-storage constraints, transferable hydrogen load constraints, interruptible hydrogen load constraints, and electric power fluctuation constraints, the first objective function and the second objective function are jointly solved to obtain the optimal new energy output consumption curve for each time period.

[0009] Furthermore, the second objective function is constructed based on the transferable hydrogen load, the shiftable hydrogen load and the interruptible hydrogen load in each time period with the goal of minimizing the hydrogen load regulation cost, including:

[0010] Based on the transferable hydrogen load, the shiftable hydrogen load and the interruptable hydrogen load in each period, the transferable hydrogen load regulation cost, the shiftable hydrogen load regulation cost and the interruptable hydrogen load regulation cost are calculated respectively;

[0011] The hydrogen load regulation cost is calculated based on the transferable hydrogen load regulation cost, the shiftable hydrogen load regulation cost and the interruptible hydrogen load regulation cost, and a second objective function is constructed with the goal of minimizing the hydrogen load regulation cost.

[0012] Furthermore, the second objective function is expressed as:

[0013]

[0014] Among them, M, I, and N are the number of transferable hydrogen load units, the number of horizontally shiftable hydrogen load units, and the number of interruptible hydrogen load units, respectively; are the transfer-out load and transfer-in load of the transferable hydrogen load of the mth unit in period t respectively; are the incentive cost coefficients of the transfer-out and transfer-in of the transferable load unit m respectively; T represents all time periods; is the incentive cost coefficient of the mobile hydrogen load unit i; is a state variable indicating whether the transferable hydrogen load is shifted to the time period τ, When indicates that the load shift of the unit starts from the τ period, When indicates that the load shift of the unit does not start from the τ period; is the output distribution vector of the i-th shiftable hydrogen load unit at the beginning of the τ period, is the unit interruptible hydrogen load cost; is the interruptible hydrogen load of the i-th interruptible hydrogen load unit in time period t.

[0015] Furthermore, based on the output consumption and predicted grid-connected output of each photovoltaic and wind power unit in each period, the first objective function is constructed with the maximum new energy consumption rate as the goal, and is expressed as:

[0016]

[0017] in, is the energy consumption rate of new energy in the T period; are the output consumption of the n1th photovoltaic and n2th wind turbines in period t, respectively; and are the predicted grid-connected output of the n1th photovoltaic and n2th wind turbines in period t; N1 and N2 are the number of photovoltaic and wind turbines, respectively.

[0018] Furthermore, based on the equivalent amount of hydrogen in the hydrogen storage tank, the total power input to the electrolyzer, the total power generation power of the equivalent electric load of the hydrogen load, the power load of the grid and the load shedding amount, the power balance constraint is constructed, which is expressed as:

[0019] D ED,t =D ED,t-1 +P t ED,in -(P t H,load -P t IL +P t sft );

[0020] P t all =P t el,load +P t ED,in -P cut,t ;

[0021] P t all =P t RE +P t non-RE ;

[0022]

[0023]

[0024] in, D ED,t-1 are the equivalent amount of hydrogen stored in the hydrogen storage tank at time t and t-1 respectively; is the total power input into the electrolytic cell during period t; are the total equivalent electric load of hydrogen load, the equivalent electric load of interruptible hydrogen load and the equivalent electric load of transferable hydrogen load during period t respectively; P t all is the total power generated by the generator in period t; P t el,load is the power load of the power grid during period t; P cut,t is the load shedding amount in period t; P t RE , P t non-RE They are the power output consumption of new energy units and traditional units respectively; are the output consumption of the n3th hydropower and n4th thermal power units in period t, respectively; N3 and N4 are the numbers of photovoltaic and wind turbines, respectively.

[0025] Furthermore, the energy storage constraint includes hydrogen energy storage constraint and electric energy storage constraint; the hydrogen energy storage constraint is constructed based on the hydrogen storage volume in the hydrogen storage tank in the previous period and the current period, the degassing / intake state of the hydrogen storage tank, the effective hydrogen storage rate of the hydrogen storage tank, and the intake and output volume of the hydrogen storage tank in the current period, and is expressed as:

[0026]

[0027] Among them, Q ED (t), Q ED (t-1) are the volumes of hydrogen stored in the hydrogen storage tank during period t and period t-1, respectively, satisfying 0≤Q ED,in (t)≤Q ED,in,max and 0≤Q ED,out (t)≤Q ED,out,max , Q ED,in,max , Q ED,out,max They represent the maximum intake and output of the hydrogen storage tank respectively; β is a 0-1 variable for the deflation / intake state, β=1 means the hydrogen storage tank is inhaled, β=0 means the hydrogen storage tank is deflated; η ED is the effective hydrogen storage rate of the hydrogen storage tank; Q ED,in (t), Q ED,out (t) are the gas intake and gas output of the hydrogen storage tank during period t.

[0028] Furthermore, the energy storage constraints are constructed based on the amount of electricity stored in the energy storage in the previous period and the current period, and the charging and discharging power of the hydrogen storage tank, which can be expressed as:

[0029]

[0030]

[0031] P ES,in (t),P ES,out (t)≤P ES,r ;

[0032] Q ES (0) = Q ES (T);

[0033] Among them, Q ES (t), Q ES (t-1) are the amount of electricity stored in the energy storage during time periods t and t-1 respectively; θ is a 0-1 variable, θ=1 indicates charging of the energy storage, and θ=0 indicates discharging of the energy storage; η ES P is the energy storage charging and discharging efficiency; ES,in (t), P ES,out (t) are the charging power and discharging power of the hydrogen storage tank during period t; are the maximum and minimum values ​​of the state of charge respectively; P ES,r is the rated power of the electric energy storage device.

[0034] Furthermore, based on the output consumption of photovoltaic, wind power, hydropower, and thermal power units, the equivalent total power load of each type of unit, the electric energy storage capacity, and the preset confidence level, the power fluctuation constraint is constructed, which is expressed as:

[0035]

[0036] Among them, Pr(·) represents the probability of event establishment; N1, N2, N3, and N4 are the number of photovoltaic, wind power, hydropower, and thermal power units, respectively; are the output consumption of the n1th photovoltaic, n2th wind power, n3th hydropower and n4th thermal power units in period t respectively; P t H,load is the equivalent total electrical load of hydrogen load during period t; P t SC is the system energy storage reserve capacity during period t; α is the preset confidence level.

[0037] Furthermore, the jointly solving the first objective function and the second objective function includes:

[0038] Solve the first objective function and the second objective function respectively to obtain the optimal solution of each objective function;

[0039] Constructing a joint objective function based on the optimal solution of each objective function, the first objective function and the second objective function;

[0040] The optimal solution of the joint objective function is obtained to obtain the optimal new energy output consumption curve for each time period.

[0041] Furthermore, based on the improved spider bee optimization algorithm, the first objective function, the second objective function and the joint objective function are solved respectively, and the solving steps include:

[0042] The time periods As unknown quantity;

[0043] Generating an initial population of wasps based on the good point set, wherein each individual position of wasps in the population corresponds to a set of initial values ​​of the unknown quantity; the initial population includes female wasps and male wasps;

[0044] Iteration starts based on the initial population. In each round of iterative optimization that reaches the convergence number:

[0045] Find the first local optimal solution of this iteration based on the female wasp;

[0046] Based on the first local optimal solution, further searching the surrounding area of ​​the first local optimal solution to determine a second local optimal solution;

[0047] Determine a nesting location based on the second local optimal solution;

[0048] Determine the location of the next generation of female and male wasps based on the nesting location and the location of the male wasps;

[0049] When the iteration ends, the second local optimal solution of the last round of iteration is the global optimal solution of the algorithm.

[0050] The present invention can achieve at least one of the following beneficial effects:

[0051] By considering the hydrogen load planning curve, the new energy consumption electricity is predicted with the goal of minimizing the consumption rate of new energy units and the hydrogen load regulation cost in the power grid system. The required data volume is small, the calculation process is simple, and the development trend of the hydrogen end can be effectively considered. Combined with the energy conversion loss, the future new energy consumption can be accurately obtained, which provides an important basis for power system planning, dispatching and management, and helps to ensure the stable operation of the power grid, reduce power generation costs, and reduce the wind and solar power abandonment rates.

[0052] By constructing a joint objective function based on the first objective function and the second objective function, the difficulty of solving the problem is reduced. The first objective function, the second objective function and the joint objective function are solved by using the improved spider bee algorithm based on the swarm intelligence algorithm. A more reasonable search space is adopted, the search speed is fast, the solution accuracy is high, and the accuracy of the prediction is improved.

[0053] Other features and advantages of the present invention will be described in the following description, and some advantages may become apparent from the description, or may be understood by practicing the present invention. The purpose and other advantages of the present invention may be realized and obtained through the contents particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the entire drawings, the same reference symbols represent the same components;

[0055] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0056] 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.

[0057] A specific embodiment of the present invention discloses a method for predicting the consumption of new energy electricity considering the cost of hydrogen load regulation, which specifically includes the following steps:

[0058] Step S1, based on the output consumption and predicted grid-connected output of each photovoltaic and wind power unit in each time period, a first objective function is constructed with the goal of maximizing the new energy consumption rate, wherein the new energy includes photovoltaic and wind power;

[0059] Step S2, based on the transferable hydrogen load, the shiftable hydrogen load and the interruptible hydrogen load in each time period, a second objective function is constructed with the goal of minimizing the hydrogen load regulation cost;

[0060] Step S3, based on the hydrogen load planning curve, power balance constraints, new energy output constraints, energy storage constraints, hydrogen production-storage constraints, transferable hydrogen load constraints, interruptible hydrogen load constraints, and electric power fluctuation constraints, jointly solve the first objective function and the second objective function to obtain the optimal new energy output consumption curve for each time period.

[0061] It should be noted that the energy units in the power grid system studied in the present invention include wind power, photovoltaic, hydropower, thermal power, electric energy storage, electrolyzer and hydrogen storage tank, and the load includes the hydrogen end and the power grid electricity load. Among them, wind power, photovoltaic, hydropower and thermal power are the power generation end units of the power grid, photovoltaic and wind power units are new energy units, hydropower and thermal power units are traditional units; electric energy storage and hydrogen storage tanks are energy storage units; electrolyzers are used to convert excess electrical energy into chemical energy, namely hydrogen energy. Electric energy storage, hydrogen storage tanks and electrolyzers work in coordination, and together with the power generation end units, they support the stable operation of the power grid and the efficient use of energy, improve the flexibility and regulation capabilities of the power grid, and promote the sustainable development of energy.

[0062] In this embodiment, in a system situation where scheduling optimization is performed with the goal of maximizing the absorption rate of new energy units (including wind power and photovoltaic units) and minimizing the hydrogen load regulation cost in the power grid system, the new energy absorption power is predicted. The required data volume is small, the calculation process is simple, and the development trend of the hydrogen end can be effectively considered. Combined with the energy conversion loss, the future new energy absorption capacity can be accurately obtained, which provides an important basis for power system planning, scheduling and management, and helps to ensure the stable operation of the power grid, reduce the cost of power generation, and reduce the wind and solar abandonment rates.

[0063] Specifically, in step S1, the first objective function is expressed as:

[0064]

[0065] in, is the energy consumption rate of new energy in period t; are the output consumption of the n1th photovoltaic and n2th wind turbines in period t, respectively, which are the parameters to be predicted; and The predicted grid-connected output of the n1th photovoltaic and n2th wind turbines in time period t are known data provided by the power grid system; N1 and N2 are the number of photovoltaic and wind turbines, respectively. T is the total number of time periods, and T time periods constitute a dispatch period. Generally, a dispatch period refers to 24 hours within a day.

[0066] Specifically, step S2 includes:

[0067] Based on the transferable hydrogen load, the shiftable hydrogen load and the interruptable hydrogen load in each period, the transferable hydrogen load regulation cost, the shiftable hydrogen load regulation cost and the interruptable hydrogen load regulation cost are calculated respectively;

[0068] The hydrogen load regulation cost is calculated based on the transferable hydrogen load regulation cost, the shiftable hydrogen load regulation cost and the interruptible hydrogen load regulation cost, and a second objective function is constructed with the goal of minimizing the hydrogen load regulation cost.

[0069] It should be noted that the hydrogen load adjustment cost is calculated based on a scheduling period, i.e., T time periods. Hydrogen load includes transferable hydrogen load and interruptible hydrogen load. Some transferable loads require the overall process flow to be shifted due to their process characteristics, which are recorded as transferable hydrogen loads.

[0070] Furthermore, the second objective function is expressed as:

[0071]

[0072] Among them, M, I, and N are the number of transferable hydrogen load units, the number of translocatable hydrogen load units, and the number of interruptible hydrogen load units, respectively, which are known parameters; are the transfer-out load and transfer-in load of the transferable hydrogen load of the mth unit in period t, which are the data to be predicted; are the incentive cost coefficients of the transfer-out and transfer-in of the transferable load unit m, which are known parameters; T represents all time periods; is the incentive cost coefficient of the mobile hydrogen load unit i, which is a known parameter; The state variable indicating whether the transferable hydrogen load is shifted to the time period τ is the data to be predicted. When indicates that the load shift of the unit starts from the τ period, When indicates that the load shift of the unit does not start from the τ period; is the output distribution vector of the i-th shiftable hydrogen load unit at the beginning of the τ period, and is the data to be predicted; is the unit interruptible hydrogen load cost, which is a known parameter; is the interruptible hydrogen load of the i-th interruptible hydrogen load unit in time period t, which is the data to be predicted.

[0073] Furthermore, when predicting the consumption of renewable energy output based on the first objective function and the second objective function, it is necessary to consider the system capacity elasticity and internal changes of the load, as well as the energy loss in the energy conversion link within the system. Therefore, it is necessary to solve the objective function based on the various constraints of the power grid.

[0074] Specifically, in step S3, an electric quantity balance constraint is constructed based on the equivalent electric quantity of hydrogen in the hydrogen storage tank, the total power input into the electrolyzer, the total electric power generation power of the equivalent electric load of the hydrogen load, the electric load of the power grid and the load shedding amount. The electric quantity balance constraint is expressed as:

[0075] D ED,t =D ED,t-1 +P t ED,in -(P t H,load -P t IL +P t sft );

[0076] P t all =P t el,load +P t ED,in -P cut,t ;

[0077] Among them, D ED,t , D ED,t-1 are the equivalent amount of hydrogen stored in the hydrogen storage tank at time t and t-1 respectively; P t ED,in is the total power input into the electrolytic cell during period t; P t H,load , P t IL , P t sft are the total hydrogen load equivalent electric load, the interruptible hydrogen load equivalent electric load and the transferable hydrogen load equivalent electric load in time period t, which are calculated based on the energy conversion coefficient of hydrogen load and electric load (the electric energy required to produce unit mass of hydrogen is calculated based on the energy conversion relationship of the electrolyzer as the energy conversion coefficient of hydrogen load and electric load); P t all is the total power generated by the generator in period t; P t el,load is the power load of the power grid during period t; P cut,t is the load shedding amount in period t, which is the known data provided by the power grid system.

[0078] Furthermore, P t all =Pt RE +P t non-RE , where P t RE , P t non-RE They are the power output consumption of new energy units and traditional units respectively.

[0079] Furthermore, in are the output consumption of the n1th photovoltaic and n2th wind turbines in period t, respectively; N1 and N2 are the number of photovoltaic and wind turbines, respectively.

[0080] Furthermore, in are the output consumption of the n3th hydropower and n4th thermal power units in period t, respectively; N3 and N4 are the numbers of photovoltaic and wind turbines, respectively.

[0081] Specifically, in step S3, the new energy output constraint is expressed as:

[0082]

[0083]

[0084] in, They are the maximum outputs of the n1th photovoltaic and n2th wind turbines in period t respectively.

[0085] Specifically, in step S3, the energy storage constraints include hydrogen energy storage constraints and electric energy storage constraints.

[0086] Furthermore, the hydrogen energy storage constraint is constructed based on the hydrogen storage volume in the hydrogen storage tank in the previous period and the current period, the degassing / intake state of the hydrogen storage tank, the effective hydrogen storage rate of the hydrogen storage tank, and the intake and output volume of the hydrogen storage tank in the current period, which is expressed as:

[0087]

[0088] Among them, Q ED (t), Q ED (t-1) are the volumes of hydrogen stored in the hydrogen storage tank during period t and period t-1, respectively, satisfying 0≤Q ED,in (t)≤Q ED,in,max and 0≤Q ED,out (t)≤Q ED,out,max , Q ED,in,max , Q ED,out,max They represent the maximum intake and output of the hydrogen storage tank respectively; β is a 0-1 variable for the deflation / intake state, β=1 means the hydrogen storage tank is inhaled, β=0 means the hydrogen storage tank is deflated; η EDis the effective hydrogen storage rate of the hydrogen storage tank; Q ED,in (t), Q ED,out (t) are the gas intake and gas output of the hydrogen storage tank during period t.

[0089] Furthermore, the energy storage constraints are constructed based on the amount of electricity stored in the energy storage in the previous period and the current period, and the charging and discharging power of the hydrogen storage tank, which can be expressed as:

[0090]

[0091]

[0092] P ES,in (t),P ES,out (t)≤P ES,r ;

[0093] Q ES (0) = Q ES (T);

[0094] Among them, Q ES (t), Q ES (t-1) are the amount of electricity stored in the energy storage during time periods t and t-1 respectively; θ is a 0-1 variable, θ=1 indicates charging of the energy storage, and θ=0 indicates discharging of the energy storage; η ES P is the energy storage charging and discharging efficiency; ES,in (t), P ES,out (t) are the charging power and discharging power of the hydrogen storage tank during period t; are the maximum and minimum values ​​of the state of charge respectively; P ES,r is the rated power of the electric energy storage device; Q ES (0), Q ES (T) are the initial storage capacity of the energy storage device and the storage capacity at the end of period T, respectively.

[0095] Specifically, in step S3, the hydrogen production-storage link includes two parts: an electrolyzer and a hydrogen storage tank, and the hydrogen production-storage constraints include electrolyzer power constraints, temperature constraints and hydrogen storage tank capacity constraints.

[0096] Furthermore, the electrolyzer power constraint is expressed as: is the minimum value of the electrolytic cell input power; P t ED is the input power of the electrolyzer during period t; The maximum value of the power input to the electrolyzer.

[0097] The electrolytic cell temperature constraint is expressed as: is the minimum working temperature of the electrolytic cell; T t EDis the operating temperature of the electrolytic cell during period t; It is the maximum operating temperature of the electrolytic cell.

[0098] The hydrogen storage tank capacity constraint is expressed as: 0≤Q ED (t)≤Q ED,max ;Q ED (t) is the volume of hydrogen stored in the hydrogen storage tank during period t; Q ED,max The maximum capacity of the hydrogen storage tank.

[0099] Specifically, in step S3, the transferable hydrogen load constraints include:

[0100]

[0101]

[0102]

[0103] Among them, P t sft The hydrogen equivalent electric load demand of the user after the transfer in time period t; is the equivalent electrical load of the transferable hydrogen load during period t; is the change in equivalent electric load of transferable hydrogen in period t; C sft The proportional coefficient of the set transferable load.

[0104] Specifically, in step S3, the interruptible hydrogen load constraint is expressed as: Where P t IL is the interruptible hydrogen equivalent electric load capacity called in period t; The maximum allowed call capacity of the system.

[0105] Specifically, in step S3, the electric power fluctuation constraint is expressed as:

[0106]

[0107] Among them, Pr(·) represents the probability of event establishment; N1, N2, N3, and N4 are the number of photovoltaic, wind power, hydropower, and thermal power units, respectively; are the output consumption of the n1th photovoltaic, n2th wind power, n3th hydropower and n4th thermal power units in period t respectively; P t H,load is the equivalent total electrical load of hydrogen load during period t; P t SC is the system electrical energy storage reserve capacity during period t; α is the preset confidence level, and the preferred setting value is above 0.95.

[0108] It should be noted that in step S03, the first objective function and the second objective function are jointly solved based on the hydrogen load planning curve and the above-mentioned multiple constraints. When solving, other constraints such as grid power supply balance constraints, unit output constraints, unit climbing output constraints and hydrogen supply balance constraints can also be considered.

[0109] Specifically, in step S3, jointly solving the first objective function and the second objective function includes:

[0110] Solve the first objective function and the second objective function respectively to obtain the optimal solution of each objective function, which is recorded as and

[0111] Based on the optimal solution of each objective function, the first objective function and the second objective function, a joint objective function is constructed. The joint objective function is:

[0112]

[0113] The optimal solution of the joint objective function is obtained to obtain the optimal new energy output consumption curve for each time period and the optimized hydrogen load planning curve.

[0114] Further, in step S3, the first objective function, the second objective function and the joint objective function are solved respectively based on the improved spider bee optimization algorithm. It should be noted that the ordinary spider bee algorithm is based on the study of the hunting, nesting and mating behaviors of female wasps in nature, assuming that the wasp larvae are specialized parasites, that is, only one egg is laid in the abdomen of each spider. The ordinary spider bee algorithm has the advantages of fast search speed and high solution accuracy. In the step of initializing the population, the improved spider bee algorithm is improved from randomly generating the initial population to generating the initial population based on the optimal point set. On the basis of the original ordinary spider bee algorithm, the generation of the initial population is further optimized, and the initial search starting point is optimized, so that it is more reasonably distributed in the search space, reducing the probability of the algorithm falling into the local optimum.

[0115] The solution process of the improved spider bee algorithm is described in detail below.

[0116] Step S31, setting the initial population.

[0117] Specifically, the As the unknown quantity in the improved spider-bee optimization algorithm.

[0118] Specifically, when solving each objective function, the range of feasible solutions for each objective function can be determined based on the hydrogen load planning curve and in combination with various constraints.

[0119] Furthermore, an initial population of wasps is generated based on the good point set, and each individual position of wasps in the population corresponds to a set of initial values ​​of the unknown quantity. The initial values ​​all satisfy the range of the above feasible solutions. It should be noted that in the process of solving each objective function based on the improved spider bee algorithm, each constraint condition must be satisfied.

[0120] The initial population generated based on the good point set is more reasonably distributed in the search space than the randomly generated initial population, which can reduce the probability of the algorithm falling into the local optimum. Furthermore, the population is divided into two parts, representing female wasps and male wasps respectively.

[0121] The mathematical form of the good point set is:

[0122]

[0123] where V=(v1,v1,...,v D ) is the optimal point vector, D is the vector dimension, corresponding to the number of unknown quantities; p is the smallest prime number that satisfies (p-3) / 2≥D; P D (i) represents the good point set; i represents the i-th individual (each individual corresponds to a set of solutions for the unknown quantity).

[0124] use Represents the initial population space, where N is the number of individuals in the population, preferably N=100.

[0125] Step S32: performing iterative solution based on the initial population.

[0126] Specifically, the population iteration is started based on the initial population, and in each generation of population optimization process that reaches the population mating convergence number, steps S321 to S325 are included:

[0127] S321. Find the first dominant area of ​​the generation based on the female wasps, and search in the first dominant area to obtain a first local optimal solution.

[0128] Specifically, step S321 in the algorithm represents "searching", that is, looking for prey at the beginning of optimization, looking for spiders suitable for larval growth, that is, conducting a preliminary search of the global area, determining the first dominant area of ​​the population of this generation, and based on the objective function currently being solved, searching in the first dominant area to obtain the first local optimal solution.

[0129] The search process is:

[0130]

[0131]

[0132]

[0133]

[0134] in, are the current search position and the next search location of the i-th female wasp individual; are the lower and upper bounds of the initial parameters in the algorithm; μ1 and μ2 are the step length ratios along the current direction; are random numbers generated using normal distribution; r1, r3, and r4 are three random numbers in the range of [0,1]; is a D-dimensional random initialization vector, and the size of each element ranges from [0,1]; l is a random number, l∈[-2,1]; Three different female wasps are randomly selected from the population to determine a current location of the exploration direction.

[0135] S322: Based on the first local optimal solution, further search the surrounding area to find a second local optimal solution.

[0136] Specifically, in the algorithm, step S322 represents "search and escape", that is, after finding the prey spider, the spider may try to escape from the center of the web, so the female wasp will follow them, select the most suitable one to paralyze and drag away. At this stage, the algorithm further searches the surrounding area of ​​the currently determined optimal solution to avoid falling into the local optimal solution, that is, based on the first local optimal solution, further search its surrounding area to find the second local optimal solution of the current solution objective function.

[0137] The search process is:

[0138]

[0139]

[0140]

[0141] Wherein, Eq. represents a formula, and an exemplary example is Eq. (2) i.e. formula (2); They are the current position of the female wasp and the location of the next search step; A current location point is used to determine the exploration direction by randomly selecting a different female wasp from the population; is a D-dimensional random initialization vector, and the size of each element is in the range of [0,1]; r6 and q are random numbers in the range of [0,1]; g t , g t,best They are respectively the objective function value of the current search position and the optimal objective function value currently searched; is a set of vectors with sizes ranging from -k to k generated according to normal distribution; k is the gradually increasing distance between the female bee and the spider.

[0142] S323, based on the second local optimal solution, find a nesting location to prepare for the generation of the next generation. Further, the second local optimal solution is the current local optimal solution of the population.

[0143] Specifically, step S323 represents nesting in the algorithm, simulating dragging the prey into a nest of a size suitable for the spider to place and lay eggs, that is, based on the second local optimal solution, finding a nesting location to prepare for the production of the next generation.

[0144]

[0145]

[0146]

[0147] in, is the optimal solution of the population so far; Three different female wasps are randomly selected from the population to determine a current location of the exploration direction; is a D-dimensional random initialization vector, and the size of each element ranges from [0,1]; χ is a flight characteristic parameter; is a binary vector used to prevent two nests from being built at the same location; N is the number of individuals in the population.

[0148] S324. Determine the positions of the next generation of female wasps and male wasps based on the nesting position and the position of the male wasp.

[0149] Specifically, step S324 represents mating in the algorithm, simulates the properties of offspring produced by mating between male and female bees with a specific probability, simulates the exchange and transmission of information, changes the population information of the next generation to search for the optimal solution again, updates the alternative solutions, avoids falling into the local optimal solution, and promotes the emergence of the global optimal solution.

[0150] The mating process is:

[0151]

[0152]

[0153]

[0154]

[0155] Among them, cross(·) is the crossover operator; SW m,t Indicates the current position of the drone; CR is the crossing rate; They are the current state and the next generation state of the female wasp, respectively; is the male wasp’s lower position vector; δ, Generate random numbers for normal distribution; For genetic factors; are 4 random individuals randomly selected from the population, and they are not equal to each other.

[0156] S325. When the population iteration ends, the optimal solution of the last generation of population is the global optimal solution of the algorithm.

[0157] Specifically, it is determined whether the number of iterations reaches the convergence number. If so, the number of iterations in S323 in the current iteration is is the optimal solution of the algorithm; if not, return to S321 for the next round of population iteration.

[0158] The present embodiment discloses a method for predicting the consumption of new energy electricity taking into account the hydrogen load regulation cost. By taking into account the hydrogen load planning curve, the consumption rate of new energy units and the minimum hydrogen load regulation cost in the power grid system are used as the goal to predict the consumption of new energy electricity. The required data volume is small, the calculation process is simple, and the development trend of the hydrogen end can be effectively considered. Combined with the energy conversion loss, the improved spider bee algorithm is used to jointly solve the objective function to obtain the future new energy consumption. This provides an important basis for power system planning, scheduling and management, and helps to ensure the stable operation of the power grid, reduce the cost of power generation, and reduce the wind and solar power abandonment rates.

[0159] 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 new energy electricity consumption prediction method considering hydrogen load regulation cost, characterized in that: The steps include: Based on the output consumption and predicted grid-connected output of each photovoltaic and wind power unit in each period, the first objective function is constructed with the maximum consumption rate of new energy as the goal. The new energy includes photovoltaic and wind power. The first objective function is expressed as: in, is the energy consumption rate of new energy in the T period; are the output consumption of the n1th photovoltaic and n2th wind turbines in period t, respectively; and are the predicted grid-connected output of the n1th photovoltaic and n2th wind turbines in period t; N1 and N2 are the number of photovoltaic and wind turbines, respectively; Based on the transferable hydrogen load, the shiftable hydrogen load and the interruptable hydrogen load in each time period, a second objective function is constructed with the minimum hydrogen load regulation cost as the goal, including: calculating the transferable hydrogen load regulation cost, the shiftable hydrogen load regulation cost and the interruptable hydrogen load regulation cost respectively based on the transferable hydrogen load, the shiftable hydrogen load and the interruptable hydrogen load in each time period; calculating the hydrogen load regulation cost based on the transferable hydrogen load regulation cost, the shiftable hydrogen load regulation cost and the interruptable hydrogen load regulation cost, and constructing the second objective function with the minimum hydrogen load regulation cost as the goal; the second objective function is expressed as: Among them, M, I, and N are the number of transferable hydrogen load units, the number of horizontally shiftable hydrogen load units, and the number of interruptible hydrogen load units, respectively; are the transfer-out load and transfer-in load of the transferable hydrogen load of the mth unit in period t respectively; are the incentive cost coefficients of the transfer-out and transfer-in of the transferable load unit m respectively; T represents all time periods; is the incentive cost coefficient of the mobile hydrogen load unit i; is a state variable indicating whether the transferable hydrogen load is shifted to the time period τ, When indicates that the load shift of the unit starts from the τ period, When indicates that the load shift of the unit does not start from the τ period; is the output distribution vector of the i-th shiftable hydrogen load unit at the beginning of the τ period, is the unit interruptible hydrogen load cost; is the interruptible hydrogen load of the i-th interruptible hydrogen load unit in time period t; Based on the hydrogen load planning curve, power balance constraints, new energy output constraints, energy storage constraints, hydrogen production-storage constraints, transferable hydrogen load constraints, interruptible hydrogen load constraints, and electric power fluctuation constraints, the first objective function and the second objective function are jointly solved to obtain the optimal new energy output consumption curve for each time period; the joint solution includes: solving the first objective function and the second objective function respectively to obtain the optimal solution of each objective function, denoted as F1 MAX and Based on the optimal solution of each objective function, the first objective function and the second objective function, a joint objective function is constructed. The joint objective function is: Find the optimal solution for the joint objective function.

2. The prediction method according to claim 1, characterized in that: The power balance constraint is constructed based on the equivalent power of hydrogen in the hydrogen storage tank, the total power input to the electrolyzer, the total power generation power of the equivalent electric load of the hydrogen load, the power load of the grid and the load shedding, which is expressed as: D ED，t =D ED，t-1 +P t ED，in -(p t H,load -P t IL +P t sft ); P t all =P t el，load +P t ED，in -P cut，t ; P t all =P t RE E+P t non-RE ; Among them, D ED，t , D ED，t-1 are the equivalent amount of hydrogen stored in the hydrogen storage tank at time t and t-1 respectively; P t ED，in is the total power input into the electrolytic cell during period t; P t H，load , P t IL , P t sft are the total equivalent electric load of hydrogen load, the equivalent electric load of interruptible hydrogen load and the equivalent electric load of transferable hydrogen load during period t respectively; P t all is the total power generated by the generator in period t; P t el,load is the power load of the power grid during period t; P cut,t is the load shedding amount in period t; P t RE , P t non-RE They are the power output consumption of new energy units and traditional units respectively; are the output consumption of the n3th hydropower and n4th thermal power units in period t, respectively; N3 and N4 are the numbers of photovoltaic and wind turbines, respectively.

3. The prediction method according to claim 2, characterized in that: The energy storage constraints include hydrogen energy storage constraints and electrical energy storage constraints; the hydrogen energy storage constraints are constructed based on the hydrogen storage volume in the hydrogen storage tank in the previous period and the current period, the degassing / intake state of the hydrogen storage tank, the effective hydrogen storage rate of the hydrogen storage tank, and the intake and output volumes of the hydrogen storage tank in the current period, and are expressed as: Among them, Q ED (t), Q ED (t-1) are the volumes of hydrogen stored in the hydrogen storage tank during period t and period t-1, respectively, satisfying 0≤Q ED ,in(t)≤Q ED ,in,max and 0≤Q ED ,out(t)≤Q ED ,out,max,Q ED ,in,max,Q ED ,out,max represent the maximum intake and output of the hydrogen storage tank respectively; β is a 0-1 variable for the deflation / intake state, β=1 represents the intake of the hydrogen storage tank, β=0 represents the deflation of the hydrogen storage tank; η ED is the effective hydrogen storage rate of the hydrogen storage tank; Q ED,in (t), Q ED,out (t) are the gas intake and gas output of the hydrogen storage tank during period t.

4. The prediction method according to claim 3, characterized in that: The energy storage constraints are constructed based on the amount of electricity stored in the energy storage in the previous period and the current period, and the charging and discharging power of the hydrogen storage tank, which can be expressed as: P ES，in (t),P ES，out (t)≤P ES，r ; Q ES (0)=Q ES (T); Among them, Q ES (t), Q ES (t-1) are the amount of electricity stored in the energy storage during time periods t and t-1 respectively; θ is a 0-1 variable, θ=1 indicates charging of the energy storage, and θ=0 indicates discharging of the energy storage; η ES P is the energy storage charging and discharging efficiency; ES,in (t), P ES,out (t) are the charging power and discharging power of the hydrogen storage tank during period t; are the maximum and minimum values ​​of the state of charge respectively; P ES,r is the rated power of the electric energy storage device.

5. The prediction method according to claim 4, characterized in that: The power fluctuation constraint is constructed based on the output consumption of photovoltaic, wind power, hydropower and thermal power units, the equivalent total power load of each type of unit, the electric energy storage capacity and the preset confidence level, which is expressed as: Among them, Pr(·) represents the probability of event establishment; N1, N2, N3, and N4 are the number of photovoltaic, wind power, hydropower, and thermal power units, respectively; are the output consumption of the n1th photovoltaic, n2th wind power, n3th hydropower and n4th thermal power units in period t respectively; P t H,load is the equivalent total electrical load of hydrogen load during period t; P t SC is the system energy storage reserve capacity during period t; α is the preset confidence level.

6. The prediction method according to claim 5, characterized in that: Based on the improved spider bee optimization algorithm, the first objective function, the second objective function and the joint objective function are solved respectively, and the solving steps include: The time periods As unknown quantity; Generating an initial population of wasps based on the good point set, wherein each individual position of wasps in the population corresponds to a set of initial values ​​of the unknown quantity; the initial population includes female wasps and male wasps; Iteration starts based on the initial population. In each round of iterative optimization that reaches the convergence number: Find the first local optimal solution of this iteration based on the female wasp; Based on the first local optimal solution, further searching the surrounding area of ​​the first local optimal solution to determine a second local optimal solution; Determine a nesting location based on the second local optimal solution; Determine the location of the next generation of female and male wasps based on the nesting location and the location of the male wasps; When the iteration ends, the second local optimal solution of the last round of iteration is the global optimal solution of the algorithm.

Images