Micro-grid optimization scheduling strategy based on flexible load

Abstract

The invention relates to a micro-grid optimization scheduling strategy based on a flexible load. The strategy specifically comprises the following steps: S1, establishing a micro-grid model and a flexible load model; s2, establishing a multi-objective optimization function based on the scheduling cost, the transaction electricity price and the minimum demand for the power of the main power grid; s3, determining constraint conditions of the micro-grid model, and scheduling shiftable loads in the flexible loads under the condition of meeting self requirements of the micro-grid; and S4, solving the established model by using a PSO algorithm, and obtaining a faster convergence rate and a better convergence effect through an improved particle swarm algorithm with conditions, thereby realizing optimal scheduling. According to the scheduling strategy, the adjustability of the flexible load can be fully excavated, so that the utilization rate of new energy power generation is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of micro-grid, more particularly to a micro-grid optimization scheduling strategy based on flexible load. BACKGROUND

[0002] With the increasing proportion of flexible load in modern power systems, demand side response plays an increasingly important role in power systems. As the basic network unit of smart grid, the efficiency and reliability of the control and operation of micro-grid are the fundamental determinants of the safety, reliability, efficiency and environmental protection of smart grid. Due to the characteristics of micro-grid that flexible load is regarded as power for unified energy management and scheduling, it is more convenient to control the classification of load than distribution network. There are two problems in the current treatment of flexible load: first, different loads in each region are not classified in detail, the dispatchable margin of flexible load in the region is not evaluated, only the scheduling potential factor is considered, the specific load type is not considered, and the dispatching control ability of micro-grid to flexible load is not fully exerted; second, the same type of flexible load is not classified and managed uniformly, and the scheduling range is small. That is, there is a technical problem that the adjustability of flexible load itself is not fully tapped to improve the utilization rate of new energy generation.

[0003] Patent No. CN117394376A discloses a residential new energy micro-grid load optimization method, which includes the following steps: step 1, modeling the residential new energy micro-grid; step 2, establishing a residential residential life load optimization model; step 3, establishing an electric vehicle charging load optimization model.

[0004] Compared with the technical solution disclosed in the prior art, the contradiction between system reliability and resident electricity consumption is relieved, and the electricity experience of residents is focused on while pursuing significant load peak clipping and valley filling benefits. The prior art also has the technical problem that the adjustability of flexible load itself is not fully tapped to improve the utilization rate of new energy generation. SUMMARY

[0005] The technical problem solved by the present application is to overcome the problems of the prior art and provide a micro-grid optimization scheduling strategy based on flexible load, which can fully tap the adjustability of flexible load itself to improve the utilization rate of new energy generation.

[0006] The object of the present application is achieved by the following technical solutions:

[0007] A micro-grid optimization scheduling strategy based on flexible load is disclosed, which specifically includes the following steps:

[0008] S1: establishing a micro-grid model and a flexible load model;

[0009] S2: a multi-objective optimization function is established based on the scheduling cost, the transaction price and the minimum demand for the main power grid power;

[0010] S3: the constraint conditions of the micro-grid model are determined, and the switchable load in the flexible load is scheduled under the condition of meeting the demand of the micro-grid itself;

[0011] S4: the model is solved by using the PSO algorithm, and the improved particle swarm algorithm with conditions is used to obtain faster convergence rate and better convergence effect, so as to realize the optimal scheduling.

[0012] Preferably, the micro-grid model in step S1 comprises:

[0013] S11: photovoltaic power generation and wind power generation model;

[0014] S12: hydrogen production and hydrogen fuel cell model.

[0015] Preferably, in step S11, the output power of the wind turbine is related to the wind speed, the swept area of the blade and the wind energy utilization coefficient; the output power of the wind turbine is:

[0016]

[0017] In the formula: P WT is the output power of the wind turbine; ρ is the air density (kg / m3); A is the swept area of the impeller (m2); C p is the wind energy utilization coefficient; V is the wind speed (m / s).

[0018] The wind energy utilization coefficient C P is related to the tip speed ratio λ of the wind turbine, and the tip speed ratio λ is:

[0019]

[0020] In the formula: n is the rotating speed of the wind wheel; R is the radius of the blade; V is the wind speed, and the unit is m / s. Under a certain wind speed, changing the tip line speed of the wind wheel will cause the tip speed ratio to change, and if the rotating speed of the blade is faster, the corresponding tip speed ratio will also be larger;

[0021] The approximate relationship between the output power of the wind turbine and the wind speed can be expressed as:

[0022]

[0023] In the formula: v is the actual wind speed, v r is the rated wind speed, P r is the rated power, v ci is the cut-in wind speed, and v co is the cut-out wind speed.

[0024] The output power P of the photovoltaic panel V As follows:

[0025]

[0026] In the formula: G C is the actual light intensity; T C is the working temperature of the photovoltaic cell; G STC and T STC are the reference values of the light intensity and the ambient temperature, respectively, taking 21000 W / m2 and 25℃, respectively; a p is the power temperature coefficient; S P is the capacity of the photovoltaic cell; f P is the power factor of the photovoltaic system.

[0027]

[0028] In the formula: T α is the ambient temperature, in ℃. For the convenience of calculation, a simplified model can be used. The power output curve under the standard condition is linearized, and the expression of the output power P V of the simplified photovoltaic panel is:

[0029] P V = η·G C ·S (6)

[0030] In the formula: P V is the actual output power; η is the conversion efficiency of the photovoltaic panel; G C is the light intensity, in kW / m2; S is the area of the photovoltaic panel, in m 2 .

[0031] Preferably, in the step S12, the hydrogen production model is a method of producing hydrogen by electrolyzing water in an electrolytic cell; and the relationship between the hydrogen production amount m2 (Nm3) and the power consumption P zq (kW) is:

[0032]

[0033] In the formula: V cell is the voltage of the electrolytic chamber, in V; F is the Faraday constant, 96485.33 C / mol; n f is the Faraday efficiency, taking 99%; k1 = 0.001; k2 = 3600RT0 / P0, wherein R, T0 and P0 are the ideal gas constant (8.314 J / (mol·K)), the standard temperature (273.15 K) and the pressure (1 bar), respectively.

[0034] Preferably, the hydrogen fuel cell model uses a proton exchange membrane fuel cell, whose power P fc (kW) and output voltage V st (V) are:

[0035] P fc = V st × I fc × k1 (8)

[0036] V st = V fc × N (9)

[0037] wherein: I fc is the output current of a single cell, A; N is the number of cells in series; V fc is the output voltage of a single cell, V. The expression of V fc is:

[0038] V fc = U nernst - U act - U ohm - U conc (10)

[0039] wherein: U nernst is the thermodynamic electromotive force, V; U act is the activation voltage loss, V; U ohm is the ohmic voltage loss, V; U conc is the concentration voltage loss, V.

[0040] The consumption amount m 2o (Nm3) of hydrogen gas in the hydrogen fuel cell is:

[0041]

[0042] wherein: N A is the Avogadro constant (6.023 x 1023); e is the electronic charge (1.6 x 10-19C); V m is the molar volume of gas (22.4 x 10-3Nm3 / mol); is the cell efficiency.

[0043] The expression of m is:

[0044]

[0045] Preferably, the step S1 further comprises:

[0046] S13: constructing a hydrogen storage device model; the hydrogen storage device can be analogous to a battery, and the charging and discharging model is:

[0047] m 2cc (t+1)=m 2cc (t)+m2(t)-m 2o (t) (13)

[0048] Where: m 2cc m1 represents hydrogen storage capacity, Nm3; m2 represents hydrogen production capacity, Nm3; m 2o The amount of hydrogen consumed, Nm3

[0049] Preferably, the flexible load includes two types: reduceable load and transferable load;

[0050] The load reduction amount is defined as follows: Let P be the load reduction amount for time period i. 1n,i (kW) is:

[0051] P 1n,i =ε 1,i ×P 1,i (14)

[0052] Where: P 1,i Let ε be the load power that can be reduced during period i, in kW; 1,i This parameter represents the degree of load reduction during the current time period i, and its value ranges from 0 to 0.5.

[0053] The total cost of load reduction compensation (in RMB) is:

[0054]

[0055] The amount of transferable load transferred out is: Let P be the load transferred out during time period i. 20on,i (kW) is:

[0056] P 20on,i =ε 20,i ×P 2,i (16)

[0057] Where: P 2,i Let ε be the available load power during time period i, in kW; 20,i Let P be the parameter controlling the load transfer rate during time period i, with a value between 0 and 1. Then, the total load transfer rate P within one day... 2oa (kW) is:

[0058]

[0059] The amount of load transferred in is: Let P be the load transfer amount during time period i. 2in,i (kW) is:

[0060] P 2in,i =ε 2i,i ×P 2aa(18)

[0061] wherein: ε 2i,i is the parameter of the load transfer-in amount in the i period, and the value is between 0 and 1. Then the total load transfer-in amount (kW) in a day is:

[0062]

[0063] According to the principle that the total load transfer-out amount is equal to the total load transfer-in amount in a day, the following can be obtained:

[0064] P 2oa = P 2ia (20)

[0065] The transfer-out and transfer-in compensation cost relationship of the transferable load is respectively:

[0066]

[0067] wherein: λ 2,i and λ 3,i are the transfer-out and transfer-in compensation coefficients in the i period, respectively.

[0068] Preferably, the multi-objective optimization function comprises: a minimum target function F1 of the large power grid supplying power to the micro grid and a total cost target function F2 of flexible load scheduling and electricity price transaction; the calculation formula of the F1 is:

[0069]

[0070] wherein: P ddwna is the total power of the large power grid supplying power to the micro grid, kW; P 3,i is the fixed load power at the i moment, kW; P w,i is the wind power at the i moment, kW; P s,i is the photovoltaic power at the i moment, kW.

[0071] Since the target function is the minimum of the one-way power supply of the large power grid to the micro grid, it needs to satisfy:

[0072] (P 1,i +P 2,i +P 3,i +P zq,i +P 2in,i )>(P w,i +P s,i +P fc,i +P 1n,i +P 2on,i ) (24)

[0073] The calculation formula of the F2 is:

[0074] F2 = min (C all,n1+C all,n2 ) = min(C1+C2+C3+C4+C5) (25)

[0075] wherein: C all,n1 is the total cost of flexible load scheduling; C all,n2 is the total cost of large grid and micro grid transaction; C4 and C5 are respectively the total sum of large grid buying price and large grid selling price, and the calculation methods are respectively:

[0076]

[0077] wherein: M dl,i and M d2,i are respectively the selling and buying price unit price of i period; P ddwi,i is the power of i period micro grid to large grid, kW; P ddwo,i is the power of i period large grid to micro grid, kW.

[0078] Since the dimensions of the large grid power supply to the micro grid target function F1 and the total cost of scheduling price transaction target function F2 are different, the two target functions are dimensionally unified before weighting processing. Considering F1 and F2, the multi-objective function after weighting processing of the dimensionally unified formula (23) and formula (25) is:

[0079]

[0080] wherein: C allo is the total cost of large grid and micro grid price transaction before scheduling; P ddwna is the total power of large grid input to micro grid before scheduling, kW; r is the weight coefficient, and the value is between 0.1 and 0.9.

[0081] Preferably, the constraint conditions include micro grid power balance constraint, wind, light, hydrogen production, fuel cell output constraint and hydrogen storage device constraint;

[0082] According to the power balance constraint principle, the power balance relationship between wind, light and other new energy and large grid output and flexible load needs to be met:

[0083] P ddw,i +P w,i +P s,i +P fc,i +P 1n,i +P 2on,i =P ddw,i +P 1,i +P 2,i +P 3,i +P zq,i +P 2in,i (29)

[0084] The power constraints for wind power, photovoltaic power, electrolytic hydrogen production, and hydrogen fuel cells are as follows:

[0085]

[0086] Where: P w,min With P w,max P s,min With P s,max P zq,min With P zq,max P fc,min With P fc,max These represent the minimum and maximum power outputs, in kW, for wind power generation, photovoltaic power generation, hydrogen production, and fuel cell power generation, respectively.

[0087] The constraints on hydrogen storage capacity for establishing a hydrogen storage device are as follows:

[0088]

[0089] Where: m 2cc,min and m 2cc,max These represent the minimum and maximum allowable capacities of the hydrogen storage device, in Nm3.

[0090] While meeting the microgrid's own needs, the loads that can be transferred into the flexible load are dispatched, and the energy storage device is charged based on the electricity price when the load is transferred or whether the microgrid's own needs are still met after the transfer.

[0091] Preferably, step S4 specifically includes: solving the established model using the PSO algorithm and obtaining a faster convergence rate and better convergence effect by using a conditional improved particle swarm algorithm.

[0092] The weight ω is expressed as:

[0093]

[0094] Where: ω start ω represents the initial inertia weight; end The inertia weight is the weight at which the number of iterations is maximized; M is the maximum number of iterations; k is the current number of iterations; to make ω(M) = ω end t takes the value 0.84.

[0095] In this algorithm, ω only changes with the number of iterations. In complex systems, there may be a mismatch between the change of ω and the current convergence result, which may lead to getting trapped in local extrema. Therefore, this technique adds a convergence result feedback loop to the solution process to determine whether ω should change during the iteration process, thereby avoiding the problem of the algorithm getting trapped in local extrema.

[0096] The CPSO algorithm judges whether the algorithm falls into a local optimum according to the fitness value and the number of iterations, and accelerates the convergence speed after jumping out, and the judgment mode is shown in formula (33):

[0097]

[0098] Wherein: gbest k The global optimal solution after the kth iteration is represented; d is a precision parameter for judging stagnation. The counter S is accumulated once every stagnation generation; when the counter satisfies formula (33), ω is updated once according to formula (32), and the counter is cleared.

[0099] The CPSO algorithm has the following beneficial effects:

[0100] The CPSO algorithm has the following beneficial effects: The CPSO algorithm has the following beneficial effects:

[0101] Figure 1 The CPSO algorithm has the following beneficial effects:

[0102] Figure 2 The CPSO algorithm has the following beneficial effects:

[0103] Figure 3 The CPSO algorithm has the following beneficial effects:

[0104] Figure 4 The CPSO algorithm has the following beneficial effects: DETAILED DESCRIPTION

[0105] In order to clearly illustrate the technical features of the present application, the present application will be described in detail below, and the accompanying drawings will be combined to illustrate the present application.

[0106] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, therefore, the protection scope of the present application is not limited by the specific embodiments disclosed below.

[0107] In addition, in the description of the present application, it needs to be understood that the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "axial", "radial", "circumferential" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined as "first", "second" can be explicitly or implicitly included one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly specified and limited.

[0108] In the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection", "fixing" and the like should be understood broadly, for example, it can be fixed connection, or detachable connection, or integral; it can be mechanical connection, or electrical connection, or communication; it can be directly connected, or indirectly connected through intermediate medium, or the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0109] In the present application, unless otherwise explicitly specified and limited, the first feature is "on" or "under" the second feature, which can be direct contact between the first and second features, or indirect contact between the first and second features through an intermediate medium. In the description of the present application, the description referring to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present application, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0110] Embodiment 1

[0111] A flexible load-based microgrid optimization scheduling strategy is disclosed, which specifically comprises the following steps:

[0112] S1: establishing a microgrid model and a flexible load model;

[0113] S2: A multi-objective optimization function is established based on scheduling cost, transaction price and minimum demand for main grid power;

[0114] S3: Constraints of the micro-grid model are determined, and the switchable load in the flexible load is scheduled under the condition of meeting the demand of the micro-grid itself;

[0115] S4: The model is solved by using the PSO algorithm, and the improved particle swarm algorithm with conditions is used to obtain faster convergence rate and better convergence effect, so as to realize the optimal scheduling.

[0116] In the embodiment, the micro-grid model includes: S11: photovoltaic power generation and wind power generation model and S12: hydrogen production and hydrogen fuel cell model.

[0117] The output power of the wind turbine is related to the wind speed, the swept area of the blade and the wind energy utilization coefficient; the output power of the wind turbine is:

[0118]

[0119] In the formula: P WT is the output power of the wind turbine; ρ is the air density (kg / m3); A is the swept area of the impeller (m2); C p is the wind energy utilization coefficient; V is the wind speed (m / s);

[0120] The wind energy utilization coefficient C P is related to the tip speed ratio λ of the wind turbine, and the tip speed ratio λ is:

[0121]

[0122] In the formula: n is the rotating speed of the wind wheel; R is the radius of the blade; V is the wind speed, and the unit is m / s; under a certain wind speed, changing the tip line speed of the wind wheel will cause the tip speed ratio to change, and if the rotating speed of the blade is faster, the corresponding tip speed ratio is also larger;

[0123] The approximate relationship between the output power of the wind turbine and the wind speed can be expressed as:

[0124]

[0125] In the formula: v is the actual wind speed, v r is the rated wind speed, P r is the rated power, v ci is the cut-in wind speed, and v co is the cut-out wind speed;

[0126] The output power P V of the photovoltaic cell panel is as follows:

[0127]

[0128] G C is the actual light intensity; T C is the operating temperature of the photovoltaic cell; G STC and T STC are the reference values of light intensity and ambient temperature, respectively, 21000 W / m2and 25℃, respectively; a p is the power temperature coefficient; S P is the capacity of the photovoltaic cell; f P is the power factor of the photovoltaic system;

[0129]

[0130] T α is the ambient temperature, in ℃; for convenience of calculation, a simplified model can be used; the power output curve under standard conditions is linearized, and the expression of the output power P V of the simplified photovoltaic panel is:

[0131] P V = η·G C ·S (6)

[0132] P V is the actual output power; η is the conversion efficiency of the photovoltaic panel; G C is the light intensity, in kW / m2; S is the area of the photovoltaic panel, in m 2 .

[0133] The hydrogen production model is a method of producing hydrogen by electrolyzing water in an electrolytic cell; the relationship between the hydrogen production amount m2(Nm3) and the power consumption P zq (kW) is:

[0134]

[0135] V cell is the voltage of the electrolytic chamber, in V; F is the Faraday constant, 96485.33 C / mol; n f is the Faraday efficiency, with a value of 99%; k1=0.001; k2=3600RT0 / P0, where R, T0and P0are the ideal gas constant (8.314 J / (mol·K)), the standard temperature (273.15 K) and the pressure (1 bar), respectively.

[0136] The hydrogen fuel cell model uses a proton exchange membrane fuel cell, and the power P fc (kW) and the output voltage V st (V) are:

[0137] Pfc = V st × I fc × k1 (8)

[0138] V st = V fc × N (9)

[0139] wherein: I fc is the output current of the single cell; N is the number of series cells; V fc is the output voltage of the single cell; wherein the expression of V fc is:

[0140] V fc = U nernst - U act - U ohm - U conc (10)

[0141] wherein: U nernst is the thermodynamic electromotive force, V; U act is the activation voltage loss, V; U ohm is the ohmic voltage loss, V; U conc is the concentration voltage loss, V;

[0142] The consumption amount m 2o of hydrogen gas in the hydrogen fuel cell (Nm3) is:

[0143]

[0144] wherein: N A is the Avogadro constant (6.023×1023); e is the electronic charge (1.6×10-19C); V m is the molar volume of gas (22.4×10-3Nm3 / mol); is the battery efficiency;

[0145] wherein the expression of m is:

[0146]

[0147] The step S1 further comprises: S13: constructing a hydrogen storage device model; the hydrogen storage device can be analogous to a storage battery, and the charging and discharging model is:

[0148] m 2cc (t+1) = m 2cc (t) + m2(t) - m 2o (t) (13)

[0149] wherein: m 2cc is the hydrogen storage amount, Nm3; m2 is the hydrogen production amount, Nm3; m2o The amount of hydrogen consumed is expressed in Nm3.

[0150] This study investigates different types of loads in microgrids and proposes an optimized dispatch strategy for microgrids based on flexible loads. This strategy fully leverages the adjustability of flexible loads and utilizes their role in grid regulation to improve the utilization rate of renewable energy generation, reduce the interaction between the microgrid and the main grid, and decrease the microgrid's dependence on the main grid.

[0151] Example 2

[0152] A microgrid optimization scheduling strategy based on flexible loads is disclosed, the scheduling strategy specifically including the following steps:

[0153] S1: Establish microgrid model and flexible load model;

[0154] S2: Establish a multi-objective optimization function based on scheduling cost, transaction electricity price, and minimizing power demand on the main grid;

[0155] S3: Determine the constraints of the microgrid model and schedule the transferable loads in the flexible loads while meeting the microgrid's own needs.

[0156] S4: Solve the established model using the PSO algorithm and obtain a faster convergence rate and better convergence effect by using a conditional improved particle swarm algorithm, thereby achieving optimized scheduling.

[0157] The difference between this embodiment and Embodiment 1 is that: flexible loads include two types: loads that can be reduced and loads that can be transferred;

[0158] The load reduction amount is defined as follows: Let P be the load reduction amount for time period i. 1n,i (kW) is:

[0159] P 1n,i =ε 1,i ×P 1,i (14)

[0160] Where: P 1,i Let ε be the load power that can be reduced during period i, in kW; 1,i This parameter represents the degree of load reduction in the current time period i, and its value ranges from 0 to 0.5.

[0161] The total cost of load reduction compensation (in RMB) is:

[0162]

[0163] The amount of transferable load transferred out is: Let P be the load transferred out during time period i. 20on,i (kW) is:

[0164] P 20on,i = ε 20,i × P 2,i (16)

[0165] wherein: P 2,i is the i-period load transfer-out power, kW; ε 20,i is the i-period parameter for controlling the load transfer-out, and the value is between 0 and 1; and the total load transfer-out P 2oa (kW) in a day is:

[0166]

[0167] The load transfer-in of the transferable load is: let P 2in,i (kW) be the i-period load transfer-in; and the total load transfer-in (kW) in a day is:

[0168] P 2in,i = ε 2i,i × P 2aa (18)

[0169] wherein: ε 2i,i is the i-period parameter for controlling the load transfer-in, and the value is between 0 and 1; and the total load transfer-in (kW) in a day is:

[0170]

[0171] According to the principle that the total load transfer-out is equal to the total load transfer-in in a day, the following can be obtained:

[0172] P 2oa = P 2ia (20)

[0173] The cost relationship formulae of the load transfer-out and the load transfer-in of the transferable load are respectively:

[0174]

[0175] wherein: λ 2,i and λ 3,i are the i-period transfer-out and transfer-in compensation coefficients respectively.

[0176] Different types of loads in the micro-grid are classified, a demand side response model based on the model of the curable and transferable load is established, a target function is established by comprehensively considering the scheduling cost, the lowest transaction electricity price and the minimum power demand of the main power grid, and an improved particle swarm algorithm is used for solving to achieve the purpose of optimal scheduling.

[0177] Embodiment 3

[0178] Disclosed is a micro-grid optimal scheduling strategy based on flexible load, which specifically comprises the following steps:

[0179] S1: Establishing a micro-grid model and a flexible load model;

[0180] S2: Establishing a multi-objective optimization function based on scheduling cost, transaction price and minimum power demand to the main grid;

[0181] S3: Determining the constraint condition of the micro-grid model, and scheduling the switchable load in the flexible load under the condition of meeting the demand of the micro-grid itself;

[0182] S4: Solving the established model by using the PSO algorithm and obtaining faster convergence rate and better convergence effect by using the improved particle swarm algorithm with conditions, so as to realize the optimal scheduling.

[0183] The difference between the embodiment and embodiment 1 is that the multi-objective optimization function includes: a minimum objective function F1 of the main grid supplying power to the micro-grid and a total cost objective function F2 of flexible load scheduling and price transaction; the calculation formula of F1 is:

[0184]

[0185] Wherein: P ddwna is the total power of the main grid supplying power to the micro-grid, kW; P 3,i is the fixed load power at i moment, kW; P w,i is the wind power at i moment, kW; P s,i is the photovoltaic power at i moment, kW;

[0186] Since the objective function is the minimum of the one-way power supply of the main grid to the micro-grid, it needs to meet:

[0187] (P 1,i +P 2,i +P 3,i +P zq,i +P 2in,i )>(P w,i +P s,i +P fc,i +P 1n,i +P 2on,i ) (24)

[0188] The calculation formula of F2 is:

[0189] F2 = min (C all,n1 +C all,n2 ) = min (C1 + C2 + C3 + C4 + C5) (25)

[0190] Wherein: C all,n1 is the total cost of flexible load scheduling; C all,n2The total cost of the large grid and the micro grid transaction; C4 and C5 are the total sum of the large grid purchase price and the total sum of the large grid selling price, and the calculation methods are respectively:

[0191]

[0192] Wherein: M dl,i and M d2,i are the i period selling and buying price unit price; P ddwi,i is the i period micro grid to the large grid power, kW; P ddwo,i is the i period large grid to the micro grid power, kW;

[0193] Because the dimension of the large grid to the micro grid power target function F1 and the dimension of the dispatching price transaction total cost target function F2 are different, the two target functions are dimensionally unified before weighting processing; considering F1 and F2, the multi-objective function after weighting addition processing of the dimensionally unified formula (23) and formula (25) is:

[0194]

[0195] Wherein: C allo is the total cost of the large grid and the micro grid price transaction before dispatching; P ddwna is the total power of the large grid input to the micro grid before dispatching, kW; r is the weight coefficient, the value is between 0.1 and 0.9.

[0196] 9. The micro grid optimal dispatching strategy based on flexible load according to claim 1, wherein the constraint conditions include micro grid power balance constraint, wind, light, hydrogen production, fuel cell output constraint and hydrogen storage device constraint.

[0197] According to the power balance constraint principle, the power balance relationship between wind, light and other new energy and large grid output and flexible load needs to be met:

[0198] P ddw,i +P w,i +P s,i +P fc,i +P 1n,i +P 2on,i =P ddw,i +P 1,i +P 2,i +P 3,i +P zq,i +P 2in,i (29)

[0199] The power constraint conditions of wind power, photovoltaic, electrolytic hydrogen production, hydrogen fuel cell are established as:

[0200]

[0201] Where: P w,min With P w,max P s,min With P s,max P zq,min With P zq,max P fc,min With P fc,max These are the minimum and maximum power outputs, in kW, for wind power generation, photovoltaic power generation, hydrogen production, and fuel cell power generation, respectively.

[0202] The constraints on hydrogen storage capacity for establishing a hydrogen storage device are as follows:

[0203]

[0204] Where: m 2cc,min and m 2cc,max These represent the minimum and maximum allowable capacities of the hydrogen storage device, in Nm3.

[0205] While meeting the microgrid's own needs, transferable loads from the flexible load pool are dispatched, and the decision to charge the energy storage device is made based on the electricity price at the time of load transfer or whether the microgrid's own needs are still met after the transfer. The specific dispatching process is as follows: Figure 2 As shown. Figure 2 Chinese: P NE P represents the total power output of wind and solar power generation, expressed in kW. L Total load power, kW; P 1n,max P 2on,max With P 2in,max These represent the maximum load reduction, maximum load transfer-out, and maximum load transfer-in, respectively, in kW.

[0206] Step S4 specifically includes: solving the established model using the PSO algorithm and obtaining a faster convergence rate and better convergence effect by using a conditional improved particle swarm algorithm.

[0207] The weight ω is expressed as:

[0208]

[0209] Where: ω start ω represents the initial inertia weight; end The inertia weight is the weight at which the number of iterations is maximized; M is the maximum number of iterations; k is the current number of iterations; to make ω(M) = ω end t takes the value 0.84;

[0210] In the algorithm, ω only changes according to the change of the iteration number, and in a complex system, the change of ω and the current convergence result may not match, thus leading to falling into a local extremum; therefore, the technology adds a convergence result feedback link in the solving process, and judges whether ω should change in the iteration process, thus avoiding the problem of falling into a local extremum;

[0211] The CPSO algorithm judges whether the algorithm falls into a local optimum and accelerates the convergence speed after jumping out according to the fitness value and the iteration number, and the judgment mode is shown in formula (33):

[0212]

[0213] wherein: gbest k gbest represents the global optimal solution after the kth iteration; d is a precision parameter for judging stagnation; the counter S is accumulated once every stagnation generation; when the counter satisfies formula (33), ω is updated once according to formula (32), and the counter is cleared.

[0214] Different types of loads in the microgrid are classified, a demand side response model based on the model of the curable and transferable load is established, a target function is established by comprehensively considering the scheduling cost, the minimum transaction electricity price and the minimum power demand of the main power grid, and the improved particle swarm algorithm is used for solving to achieve optimal scheduling.

[0215] Obviously, the above embodiments are only examples for clearly illustrating the technical solutions of the present application, and are not intended to limit the implementation manners of the present application. For those skilled in the art, on the basis of the above description, other different forms of changes or modifications can also be made. Here, it is not necessary and impossible to exhaust all the implementation manners. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the claims of the present application.

Claims

1. A microgrid optimized scheduling strategy based on flexible loads, characterized in that, The scheduling strategy specifically includes the following steps: S1: Establish microgrid model and flexible load model; S2: Establish a multi-objective optimization function based on scheduling cost, transaction electricity price, and minimizing power demand on the main grid; S3: Determine the constraints of the microgrid model and schedule the transferable loads in the flexible loads while meeting the microgrid's own needs. S4: Solve the established model using the PSO algorithm and obtain a faster convergence rate and better convergence effect by using a conditional improved particle swarm algorithm, thereby achieving optimized scheduling.

2. The microgrid optimization scheduling strategy based on flexible loads according to claim 1, characterized in that, The microgrid model in step S1 includes: S11: Photovoltaic power generation and wind power generation models; S12: Hydrogen production and hydrogen fuel cell model.

3. The microgrid optimization scheduling strategy based on flexible loads according to claim 2, characterized in that, In step S11, the output power of the wind turbine is related not only to the wind speed, but also to the area swept by the blades and the wind energy utilization coefficient; the output power of the wind turbine is: In the formula: P WT ρ is the output power of the wind turbine; A is the air density; C is the swept area of ​​the impeller; ρ is the air density ... p V is the wind energy utilization coefficient; V is the wind speed. Wind energy utilization coefficient C P The tip speed ratio λ of the wind turbine blades is related to the tip speed ratio λ, which is: In the formula: n is the rotor speed; R is the blade radius; V is the wind speed, in m / s; At a certain wind speed, changing the tip speed of the wind turbine blades will cause the tip speed ratio to change. If the blade rotation speed increases, the corresponding tip speed ratio will also increase. The approximate relationship between the output power of a wind turbine and wind speed can be expressed as: In the formula: v is the actual wind speed, v r For the rated wind speed, P r For rated power, v ci To cut off the wind speed, v co To cut off the wind speed; The output power P of the photovoltaic panel V as follows: In the formula: G C This represents the actual light intensity. T C The operating temperature of the photovoltaic cell; G STC and T STC These are reference values ​​for light intensity and ambient temperature, respectively, taken as 21000 W / m² and 25℃; α p S is the power temperature coefficient; P For photovoltaic cell capacity; f P The power factor of the photovoltaic system; In the formula: T α This refers to the ambient temperature, measured in °C. For ease of calculation, a simplified model can be used. The power output curve under standard conditions is linearized, and the simplified output power P of the photovoltaic panel is obtained. V The expression: P V =η·G C ·S In the formula: P V η is the actual output power; η is the conversion efficiency of the photovoltaic panel; G C S represents the light intensity, measured in kW / m²; S represents the area of ​​the photovoltaic panel, measured in m². 2 .

4. The microgrid optimization scheduling strategy based on flexible loads according to claim 2, characterized in that, In step S12, the hydrogen production model is based on the method of producing hydrogen gas by electrolyzing water in an electrolyzer; let the hydrogen production rate m2 (Nm3) and the power consumption P be... zq The relationship for (kW) is: In the formula: V cell V is the voltage of the electrolysis chamber; F is the Faraday constant, 96485.33 C / mol; n f The Faraday efficiency is taken as 99%; k1 = 0.001; k2 = 3600RT0 / P0, where R, T0 and P0 are the ideal gas constant (8.314 J / (mol·K)), standard temperature (273.15 K) and pressure (1 bar), respectively.

5. A microgrid optimization scheduling strategy based on flexible loads according to claim 2, characterized in that, The hydrogen fuel cell model described uses a proton exchange membrane fuel cell with a power output P. fc (kW) and output voltage V st (V) is: P fc =V st ×I fc ×k1 V st =V fc ×N Among them: I fc V represents the output current of a single battery cell; N represents the number of batteries connected in series; V fc V is the output voltage of a single battery cell; where V fc The expression is: V fc =U nernst -U act -U ohm -U conc Among them: U nernst Let V be the sterning electromotive force; U be the sterning electromotive force. act For activation voltage loss, V; U ohm For ohmic voltage loss, V; U conc For concentration voltage loss, V; Hydrogen consumption m in a hydrogen fuel cell 2o (Nm3) is: Where: N A is Avogadro's constant (6.023 × 10²³); e is the electron charge (1.6 × 10⁻¹⁹ C); V m The molar volume of the gas is 22.4 × 10⁻³ Nm³ / mol. For battery efficiency; in The expression is:

6. The microgrid optimization scheduling strategy based on flexible loads according to claim 2, characterized in that, Step S1 further includes: S13: Construct a model for a hydrogen storage device; the hydrogen storage device can be compared to a battery, and the charge / discharge model is as follows: m 2cc (t+1)=m 2cc (t)+m2(t)-m 2o (t) Where: m 2cc m1 represents hydrogen storage capacity; m2 represents hydrogen production capacity; m 2o This represents the amount of hydrogen consumed.

7. The microgrid optimization scheduling strategy based on flexible loads according to claim 1, characterized in that, The flexible load includes two types: load that can be reduced and load that can be transferred. The load reduction amount is defined as follows: Let P be the load reduction amount for time period i. 1n,i (kW) is: P 1n,i =e 1,i ×P 1,i Where: P 1,i Let ε be the load power that can be reduced during period i, in kW; 1,i This parameter represents the degree of load reduction in the current time period i, and its value ranges from 0 to 0.

5. The total cost of load reduction compensation (in RMB) is: The amount of transferable load transferred out is: Let P be the load transferred out during time period i. 20on,i (kW) is: P 20on,i =e 20,i ×P 2,i Where: P 2,i Let ε be the available load power during time period i, in kW; 20,i Let P be the parameter for controlling the load transfer amount during time period i, with a value between 0 and 1; then the total load transfer amount P within a day. 2oa (kW) is: The amount of load transferred in is: Let P be the load transfer amount during time period i. 2in,i (kW) is: P 2in,i =e 2i,i ×P 2aa Where: ε 2i,i Let i be the parameter for controlling the load transfer in time period i, with a value between 0 and 1; then the total load transfer in a day (kW) is: Based on the principle that the total load outflow equals the total load inflow within a day, we can conclude that: P 2oa =P 2ia The formulas for the transfer-out and transfer-in compensation costs of the transferable load are as follows: Where: λ 2,i and λ 3,i These are the transfer-out and transfer-in compensation coefficients for time period i, respectively.

8. The microgrid optimization scheduling strategy based on flexible loads according to claim 1, characterized in that, The multi-objective optimization function includes: an objective function F1 for minimizing power supply from the large power grid to the microgrid, and an objective function F2 for the total cost of flexible load dispatching and electricity price trading; the formula for calculating F1 is: Where: P ddwna The total power supplied by the large power grid to the microgrid, in kW; P 3,i Let P be the fixed load power at time i, in kW; w,i Let P be the wind power generation capacity at time i, in kW; s,i Let i be the photovoltaic power generation at time i, in kW; Since the objective function is to minimize the unidirectional power supply from the large power grid to the microgrid, it must satisfy: (P 1,i +P 2,i +P 3,i +P zq,i +P 2in,i )>(P w,i +P s,i +P fc,i +P 1n,i +P 2on,i ) The formula for calculating F2 is: F2=min(C all,n1 +C all,n2 )=min(C1+C2+C3+C4+C5) Where: C all,n1 C represents the total cost of flexible load dispatching. all,n2 C4 represents the total transaction cost between the large power grid and the microgrid; C5 and C4 represent the total purchase price and the total sale price of electricity from the large power grid, respectively, calculated using the following methods: Where: M dl,i and M d2,i These represent the unit prices for electricity sold and purchased during time period i; P ddwi,i P represents the power transmitted from the microgrid to the main grid during time period i, in kW. ddwo,i Let i be the power transmitted from the large power grid to the microgrid during time period i, in kW; Because the objective functions F1 (supply of electricity from the large power grid to the microgrid) and F2 (total cost of dispatching electricity price transactions) have different dimensions, the dimensions of the two objective functions were unified before weighting. The multi-objective function after weighted summation is as follows: Where: C allo The total cost of electricity price transactions between the large power grid and microgrids before dispatching; P ddwna The total power input to the microgrid from the main power grid before dispatch is kW; r is the weighting coefficient, with a value between 0.1 and 0.

9.

9. A microgrid optimization scheduling strategy based on flexible loads according to claim 1, characterized in that, The constraints include microgrid power balance constraints, wind, solar, hydrogen production, fuel cell output constraints, and hydrogen storage device constraints. According to the power balance constraint principle, the power balance relationship between new energy sources such as wind and solar power and the output of the main power grid and flexible loads must be satisfied: P ddw,i +P w,i +P s,i +P fc,i +P 1n,i +P 2on,i =P ddw,i +P 1,i +P 2,i +P 3,i +P zq,i +P 2in,i The power constraints for wind power, photovoltaic power, electrolytic hydrogen production, and hydrogen fuel cells are as follows: Where: P w,min With P w,max P s,min With P s,max P zq,min With P zq,max P fc,min With P fc,max These are the minimum and maximum power outputs, in kW, for wind power generation, photovoltaic power generation, hydrogen production, and fuel cell power generation, respectively. The constraints on hydrogen storage capacity for establishing a hydrogen storage device are as follows: Where: m 2cc,min and m 2cc,max These represent the minimum and maximum allowable capacities of the hydrogen storage device, in Nm3. While meeting the microgrid's own needs, the loads that can be transferred into the flexible load are dispatched, and the energy storage device is charged based on the electricity price when the load is transferred or whether the microgrid's own needs are still met after the transfer.

10. A microgrid optimization scheduling strategy based on flexible loads according to claim 1, characterized in that, Step S4 specifically includes: solving the established model using the PSO algorithm and obtaining a faster convergence rate and better convergence effect by using a conditional improved particle swarm algorithm. The weight ω is expressed as: Where: ω start ω represents the initial inertia weight; end The inertia weight is the weight at which the number of iterations is maximized; M is the maximum number of iterations; k is the current number of iterations; to make ω(M) = ω end t takes the value 0.84; In this algorithm, ω only changes with the number of iterations. In complex systems, there may be a mismatch between the change of ω and the current convergence result, which may lead to getting trapped in local extrema. Therefore, this technique adds a convergence result feedback loop to the solution process to determine whether ω should change during the iteration process, thereby avoiding the problem of the algorithm getting trapped in local extrema. The CPSO algorithm determines whether it has gotten stuck in a local optimum based on the fitness value and the number of iterations, and accelerates the convergence speed after escaping. The determination method is shown in the following formula: Among them: gbest k denoted as the global optimal solution after the k-th iteration; d is the precision parameter for determining stagnation; the counter S is accumulated once for each stagnation generation; when the counter satisfies the CPSO algorithm judgment formula, ω is updated once and the counter is cleared.

Images