Photovoltaic bearing capacity optimization method and system based on multi-element flexible load

Abstract

The invention discloses a photovoltaic bearing capacity optimization method and system based on multi-element flexible loads, and the method comprises the steps: constructing a photovoltaic bearing capacity optimization model with the minimum multi-element flexible loads as a target, and enabling the multi-element flexible loads to be divided into transferable loads, translational loads and reducible loads according to the scheduling management mode of the loads, by processing response behaviors of different flexible loads to scheduling instructions in a distinguished manner, an operation optimization strategy can be fully considered in planning, so that the multi-target opportunity constraint planning model obtains the maximum distributed photovoltaic bearing capacity, and meanwhile, the investment and operation cost of cooperation of different promotion means is taken into consideration. According to the method, a photovoltaic bearing capacity optimization scheme can consider the maximum distributed photovoltaic bearing capacity and the investment and operation cost by distinguishing multi-element flexible loads, and the solving difficulty can be effectively reduced by optimizing the solving steps through a coevolution algorithm of an improved initial population generation strategy based on LHS-MC embedding.

Description

Technical Field

[0001] The present invention relates to a photovoltaic carrying capacity optimization method, and in particular to a photovoltaic carrying capacity optimization method and system based on multi-element flexible loads. Background Art

[0002] Dense access of distributed photovoltaic power to distribution networks is a structural form of future distribution networks. The access of a large number of distributed photovoltaic power plants has changed the distribution of distribution network trends and increased the risk of voltage exceeding the limit in distribution networks. Therefore, improving the carrying capacity of distributed photovoltaic power plants in distribution networks is a key technical issue in the planning and operation of future distribution networks. In addition, the core of distribution network operation is power balance, that is, the topological structure of the distribution network is determined according to the load demand in the planning year to ensure the operational reliability of the system. However, with the increase in the proportion of access to renewable energy and electric vehicles, the system's demand is not just a simple load, and the demand for flexibility has gradually become prominent, and has far exceeded the flexibility demand of traditional power systems. It is necessary to optimize the balance and matching of load flexibility. Among them, flexible loads, as an important component of the power system, play an increasingly prominent role in improving the flexibility of the power system.

[0003] The main steps of existing flexible load scheduling include determining the research object and demand response project type, user group clustering analysis based on electricity consumption characteristics, identification of participation rate of classified demand response projects, price elasticity calculation and demand response potential assessment, with an emphasis on price elasticity calculation methods suitable for segmented user groups.

[0004] However, there are currently few studies that systematically and clearly classify flexible loads and optimize the load flexibility of power systems by analyzing multivariate models of flexible loads to improve the photovoltaic carrying capacity of distribution networks.

[0005] Although this method is feasible, it still has the following drawbacks:

[0006] 1. At present, the analysis of flexible loads is focused on the user's perspective. Due to the wide extension of flexible loads, there are many types of flexible loads, and different flexible loads have differentiated mechanism characteristics. The perspective of typical flexible resource flexible loads is not considered, resulting in poor optimization of photovoltaic carrying capacity.

[0007] 2. Due to the complex structure of the distribution network, the number of variables in the established optimization model is huge. Among them, the variables to be optimized include not only flexible loads, but also a large number of discrete variables of network branches. Ordinary optimization solution algorithms may encounter the "curse of dimensionality" problem, which makes the model solution complicated or even impossible to solve.

[0008] The information disclosed in this background technology section is only intended to increase the understanding of the overall background of the application, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the Invention

[0009] The object of the present invention is to overcome the disadvantages of poor optimization effect of photovoltaic carrying capacity and complex model solution in the prior art, and provide a photovoltaic carrying capacity optimization method based on multiple flexible loads with better optimization effect of photovoltaic carrying capacity and simple model solution.

[0010] To achieve the above object, the technical solution of the present invention is:

[0011] A photovoltaic carrying capacity optimization method based on multiple flexible loads, the optimization method includes:

[0012] S1. Construct a photovoltaic carrying capacity optimization model. The photovoltaic carrying capacity optimization model aims to minimize the multiple flexible loads. The multiple flexible loads include transferable loads, shiftable loads, and curtailable loads, and consider the constraints affecting the flexible loads.

[0013] S2. Solve the photovoltaic carrying capacity optimization model. Solve the photovoltaic carrying capacity optimization model by an improved co-evolution algorithm to obtain a photovoltaic carrying capacity optimization scheme.

[0014] The transferable load is a load with a constant total power consumption within the scheduling period but flexible adjustment of power consumption in each period. The shiftable load is a load restricted by the production process and can only achieve a large-time period shift of the power consumption curve. The curtailable load is a load whose power consumption can be reduced or interrupted through demand analysis.

[0015] The photovoltaic carrying capacity optimization model includes an in-schedulable load model and a schedulable load model. The in-schedulable load model includes:

[0016] ΔD i trans = f1(D 0i , Δp i , ε ij , v i trans );

[0017]

[0018] ΔD i shift = f2(i + Δi(Δp i )) - f2(i);

[0019] ΔD i re = f3(D 0i , Δp i , ε ii , v i re );

[0020] In the above formula, ΔD i trans is i the response volume of the shiftable load in the time period, D 0i is the reference load in the i-th time period, △p i is the electricity price difference vector between the i-th time period and other time periods, ε ij is the cross-elasticity vector of the i-th time period relative to the j-th time period, v i trans is the load transfer rate in the i-th time period, ΔD i shift is i the response volume of the shiftable load in the time period, T is the scheduling period, ΔD i re is the response volume of the load that can be curtailed in the i-th time period, ε ii is the self-elasticity vector, v i re is the load curtailment rate;

[0021] The non-schedulable load model includes a load curtailment model, a shiftable load model, and a shiftable load model. The load curtailment model includes:

[0022]

[0023]

[0024] In the above formula, is the power of the i-th load that can be curtailed in the t time period after participating in the demand response; μ i,t indicates whether the i-th load that can be curtailed is curtailed in the t time period, μ i,t =1 means that the i-th load that can be curtailed is curtailed at time period; α i is the curtailment rate of the i-th load that can be curtailed, 0 < α i < 1; is the power of the i-th load that can be curtailed in the t time period before participating in the demand response; is the capacity cost compensation of the i-th load that can be curtailed; c i is the reported unit capacity cost of the i-th load that can be curtailed; represents the operating cost of the i-th load that can be curtailed; is the electricity price cost compensation of the i-th load that can be curtailed; T represents the scheduling period;

[0025] The shiftable load model includes:

[0026]

[0027] In the above formula, $C_{i}^{shift}$ is the operating cost of the $i$-th shiftable load; $c_{i}^{shift}$ is the cost per unit power of the $i$-th shiftable load; $y_{i}^{shift}$ i,τ $y_{i}^{shift}$ is the status of the $i$-th shiftable load. If $y_{i}^{shift}$ i,τ $= 1$, it means the load starts from the $\tau$ period; $T_{i}^{shift}$ is the interval of the shiftable period acceptable for the load; $t_{start}^{shift}$ is the starting time period of the load; $P_{i}^{shift}$ is the sum of the electrical powers of the $i$-th shiftable load; $t$ Di $t_{total}^{shift}$ is the total time period of electricity consumption of the $i$-th shiftable load. The total load includes $S_{i}^{shift}$ is the set of starting periods of the $i$-th shiftable load;

[0028] The shiftable load model includes:

[0029]

[0030] In the above formula, $C_{i}^{transfer}$ is the operating cost of the $i$-th transferable load; $c_{i}^{transfer}$ is the cost per unit power of the $i$-th transferable load; $P_{i}^{transfer}$ is the sum of the electrical powers of the $i$-th transferable load.

[0031] The curtailable load constraints include:

[0032] The curtailment duration constraints for curtailable loads include a minimum duration constraint and a maximum duration constraint. The minimum duration constraint includes:

[0033]

[0034] In the above formula, $\mu_{i}^{curtail}$ i,t indicates whether the $i$-th curtailable load is curtailed in the $t$ period. $\mu_{i}^{curtail}$ i,t $= 1$ means the $i$-th curtailable load is curtailed in the $t$ period. $F_{i}^{curtail}$ i is the minimum curtailment time of the $i$-th curtailable load, is the minimum curtailment time of the $i$-th curtailable load, $t_{start}^{curtail}$ is the starting curtailment time of the $i$-th curtailable load;

[0035] The maximum duration constraint includes:

[0036]

[0037] In the above formula, $G_{i}^{curtail}$ i is the maximum curtailment time of the $i$-th curtailable load;

[0038] The shiftable load constraints include:

[0039]

[0040] In the above formula, y i,τ indicates whether the i-th translatable load starts from the τ period. y i,τ = 1 means the load starts from the τ period; is the set of start periods of the i-th translatable load;

[0041] The transferable load constraints include minimum duration constraint, load power constraint, transferable interval constraint and power balance constraint. The minimum duration constraint includes:

[0042]

[0043] In the above formula, represents the minimum duration of the i-th transferable load, and v i,τ indicates whether the i-th transferable load starts from the τ period. v i,τ = 1 means the load starts from the τ period; is the left limit point of the time interval of the i-th transferable load, is the right limit point of the time interval of the i-th transferable load.

[0044] The load power constraint includes:

[0045]

[0046] In the above formula, and respectively represent the upper and lower limits of the power of the i-th transferable load, is the power of the i-th transferable load at time t;

[0047] The power balance constraint includes:

[0048]

[0049] In the above formula, W i * represents the sum of the electricity quantities of the i-th transferable load before being called, and Δt is the operation time of the transferable load.

[0050] The S2 includes:

[0051] S2.1. Divide the control variables of the optimization model into three populations of curtailable load, translatable load and transferable load according to the type of control quantity to be optimized. The operation cost component X1 of the curtailable load = [C1 re , C2 re , …, C i re ; The operation cost component X2 of the translatable load = [C1 shift, C2 shift , …, C i shift ; The transferable load operation cost component X3 = [C1 trans , C2 trans , …, C i trans ;

[0052] S2.2. Generate initial population individuals according to the improved initial population generation strategy. The expression of the improved initial population generation strategy is:

[0053]

[0054] In the above formula, is the initial individual of the s-th population, s = 1, 2, …, S size ; r s is a random number uniformly distributed on [0, 1] for the s-th initial individual; A and B are the lower and upper limit value sets of the control variables respectively;

[0055] S2.3. Use the LHS method to sample the flexible load power / light intensity random variables to obtain D independent random samples;

[0056] S2.4. Perform probabilistic power flow calculation on the distribution network to check the feasibility of individuals in the initial population. Successively check whether the state variables of D independent random samples exceed the limits, and judge whether the individuals meet the chance constraint confidence level requirements according to the law of large numbers. If not, perform contraction and approach according to the following formula until it becomes a feasible individual:

[0057]

[0058] In the above formula, α is the contraction coefficient, with an initial value of 0.5, and the value of α is halved in each contraction process, and finally it can always make become a feasible initial individual;

[0059] S2.5. Repeat S2.2 to S2.4 until S size initial populations that meet the chance constraints are generated;

[0060] S2.6. Perform selection, crossover, and mutation operations on each sub-population according to the genetic algorithm process to generate new individuals Px'. The new individual Px' is the node power component of a group of flexible loads with the largest fitness function value;

[0061] S2.7. Inter-population cooperation. Select the individual with the highest fitness value from the other population, decode it and jointly form the control variable Px” of the system with the individual Px' within its own population, and calculate the fitness of the individual Px';

[0062] S2.8. Determine whether the termination condition for overall collaborative optimization is met. The termination condition is reaching the maximum genetic algebra k. max When the termination condition is not met, enter S2.6 for re-optimization calculation. When the termination condition is met, terminate the optimization process to obtain the Pareto optimal solution set and enter S2.9.

[0063] S2.9. Use the fuzzy method based on information entropy to select the optimal compromise solution from the obtained Pareto optimal solution set, and finally obtain a set of optimal multi-flexible load combinations to achieve the improvement of photovoltaic carrying capacity.

[0064] A photovoltaic carrying capacity optimization system based on multi-flexible loads, the system includes a model construction module and a model solution module.

[0065] The model construction module is used to construct a photovoltaic carrying capacity optimization model. The photovoltaic carrying capacity optimization model aims to minimize the multi-flexible load. The multi-flexible load includes transferable load, shiftable load, and curtailable load, and considers the constraints affecting the flexible load.

[0066] The model solution module is used to solve the photovoltaic carrying capacity optimization model through an improved co-evolution algorithm to obtain a photovoltaic carrying capacity optimization plan.

[0067] The transferable load is a load with a constant total power consumption within the scheduling period, but the power consumption in each time period can be flexibly adjusted. The shiftable load is a load restricted by the production process and can only achieve a large-time period translation of the power consumption curve. The curtailable load is a load whose power consumption can be reduced or interrupted through demand analysis.

[0068] The photovoltaic carrying capacity optimization model includes an uncontrollable load model and a controllable load model. The uncontrollable load model includes:

[0069] ΔD i trans = f1(D 0i ,Δp i ,ε ij ,v i trans );

[0070]

[0071] ΔD i shift = f2(i + Δi(Δp i )) - f2(i);

[0072] ΔD i re = f3(D 0i ,Δp i, ε ii , v i re );

[0073] In the above formula, ΔD i trans is i the response volume of the shiftable load in the time period, D 0i is the reference load in the i-th time period, △p i is the electricity price difference vector between the i-th time period and other time periods, ε ij is the cross-elasticity vector of the i-th time period relative to the j-th time period, v i trans is the load transfer rate in the i-th time period, ΔD i shift is i the response volume of the shiftable load in the time period, T is the scheduling period, ΔD i re is the response volume of the load that can be curtailed in the i-th time period, ε ii is the self-elasticity vector, v i re is the load curtailment rate;

[0074] The non-schedulable load model includes a load curtailment model, a shiftable load model, and a shiftable load model. The load curtailment model includes:

[0075]

[0076] In the above formula, is the power of the i-th type of load that can be curtailed at time t after participating in the demand response; μ i,t indicates whether the i-th type of load that can be curtailed is curtailed at time t, μ i,t = 1 means that the i-th type of load that can be curtailed is curtailed at time t; α i is the curtailment rate of the i-th type of load that can be curtailed, 0 < α i < 1; is the power of the i-th type of load that can be curtailed at time t before participating in the demand response; is the capacity cost compensation for the i-th type of load that can be curtailed; c i is the reported unit capacity cost of the i-th type of load that can be curtailed; represents the operating cost of the i-th type of load that can be curtailed; is the electricity price cost compensation for the i-th type of load that can be curtailed; T represents the scheduling period;

[0077] The shiftable load model includes:

[0078]

[0079] In the above formula, is the operating cost of the i-th shiftable load; is the cost per unit power of the i-th shiftable load; y i,τ is the status of the i-th shiftable load. If y i,τ = 1, it means the load starts from the τ period; is the interval of shiftable periods acceptable to the load; is the starting time period of the load; is the sum of the electrical powers of the i-th shiftable load; t Di is the total time period of electricity consumption of the i-th shiftable load. The total load includes is the set of starting periods of the i-th shiftable load;

[0080] The shiftable load model includes:

[0081]

[0082] In the above formula, is the operating cost of the i-th shiftable load; is the cost per unit power of the i-th shiftable load; is the sum of the electrical powers of the i-th shiftable load.

[0083] The load curtailment constraint includes:

[0084] The curtailment duration constraint for the load curtailment includes a minimum duration constraint and a maximum duration constraint. The minimum duration constraint includes:

[0085]

[0086]

[0087] In the above formula, μ i,t indicates whether the i-th load curtailment is curtailed in the t period. μ i,t = 1 means the i-th load curtailment is curtailed in the t period. F i is the minimum curtailment time of the i-th load curtailment, is the minimum curtailment time of the i-th load curtailment, is the starting curtailment time of the i-th load curtailment;

[0088] The maximum duration constraint includes:

[0089]

[0090] In the above formula, G i is the maximum curtailment time of the i-th load curtailment;

[0091] The translatable load constraint includes:

[0092]

[0093] In the above formula, y i,τ indicates whether the i-th translatable load starts from the τ period. y i,τ = 1 means the load starts from the τ period; is the set of start periods of the i-th translatable load;

[0094] The transferable load constraint includes a minimum duration constraint, a load power constraint, a transferable interval constraint, and an electric energy balance constraint. The minimum duration constraint includes:

[0095]

[0096] In the above formula, represents the minimum duration of the i-th transferable load, and v i,τ indicates whether the i-th transferable load starts from the τ period. v i,τ = 1 means the load starts from the τ period; is the left limit point of the time interval of the i-th transferable load, is the right limit point of the time interval of the i-th transferable load.

[0097] The load power constraint includes:

[0098]

[0099] In the above formula, and respectively represent the upper and lower limits of the power of the i-th transferable load, is the power of the i-th transferable load at the t time period;

[0100] The electric energy balance constraint includes:

[0101]

[0102] In the above formula, W i * represents the sum of the electric energy of the i-th transferable load before being called, and Δt is the operating time of the transferable load.

[0103] The specific operation steps of the model solving module are as follows:

[0104] S2.1. Classify the control variables of the optimization model into three populations: curtailable load, translatable load, and transferable load according to the type of control quantity to be optimized. The operating cost component of the curtailable load X1 = [C1 re , C2 re , …, Ci re ; The translatable load operation cost component X2 = [C1 shift , C2 shift , …, C i shift ; The transferable load operation cost component X3 = [C1 trans , C2 trans , …, C i trans ;

[0105] S2.2. Generate the initial population individuals according to the improved initial population generation strategy. The expression of the improved initial population generation strategy is:

[0106]

[0107] In the above formula, is the initial individual of the s-th population, s = 1, 2, …, S size ; r s is a random number uniformly distributed on [0, 1] for the s-th initial individual; A and B are the sets of lower and upper limit values of the control variables respectively;

[0108] S2.3. Use the LHS method to sample the flexible load power / light intensity random variables to obtain D independent random samples;

[0109] S2.4. Conduct probabilistic power flow calculation on the distribution network to check the feasibility of the individuals in the initial population. Successively check whether the state variables of the D independent random samples exceed the limits, and judge whether the individuals meet the opportunity constraint confidence level requirements according to the law of large numbers. If not, perform contraction and convergence according to the following formula until they become feasible individuals:

[0110]

[0111] In the above formula, α is the contraction coefficient, with an initial value of 0.5, and the value of α is halved in each contraction process, and finally will become a feasible initial individual;

[0112] S2.5. Repeat S2.2 to S2.4 until S size initial populations that meet the opportunity constraints are generated;

[0113] S2.6. Perform selection, crossover, and mutation operations on each sub-population according to the genetic algorithm process to generate new individuals Px'. The new individual Px' is the node power component of a group of flexible loads with the largest fitness function value;

[0114] S2.7, Inter-population collaboration: Select the individual with the highest fitness value from the other population. After decoding, it forms the control variable Px” of the system together with the individual Px' within its own population, and calculate the fitness of the individual Px'.

[0115] S2.8, Determine whether the termination condition for overall collaborative optimization is met. The termination condition is reaching the maximum genetic generation k. max , When the termination condition is not met, enter S2.6 for re-optimization calculation. When the termination condition is met, terminate the optimization process to obtain the Pareto optimal solution set and enter S2.9.

[0116] S2.9, Use the fuzzy method based on information entropy to select the optimal compromise solution from the obtained Pareto optimal solution set, and finally obtain a set of optimal multi-flexible load combinations to achieve the improvement of photovoltaic carrying capacity.

[0117] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0118] 1. In the method for optimizing photovoltaic carrying capacity based on multi-flexible load of the present invention, a photovoltaic carrying capacity optimization model with the minimum multi-flexible load as the goal is constructed. The multi-flexible load is divided into shiftable load, shiftable load and curtailable load according to the scheduling management method of the load. By differentiating the response behaviors of different flexible loads to the scheduling instructions, the operation optimization strategy can be fully considered in the planning, so that the multi-objective chance-constrained programming model can take into account the investment operation costs of the coordinated cooperation of different improvement means while obtaining the maximum distributed photovoltaic carrying capacity. Therefore, this design can make the photovoltaic carrying capacity optimization scheme take into account the maximum distributed photovoltaic carrying capacity and investment operation costs by differentiating the multi-flexible load, and effectively improve the optimization effect of photovoltaic carrying capacity.

[0119] 2. In the method for optimizing photovoltaic carrying capacity based on multi-flexible load of the present invention, the photovoltaic carrying capacity optimization model is solved by the co-evolution algorithm. The co-evolution algorithm embeds and improves the initial population generation strategy based on LHS-MC. For multi-objective problems, by improving the Pareto optimal solution algorithm, the co-evolution algorithm with the improved initial population generation strategy is used to obtain the Pareto solution set of multiple objectives. Finally, the optimal compromise solution is selected from the Pareto front based on the fuzzy theory. Therefore, this design can optimize the solution steps through the co-evolution algorithm based on LHS-MC embedding and improving the initial population generation strategy, and effectively reduce the solution difficulty. BRIEF DESCRIPTION OF THE DRAWINGS

[0120] Figure 1 is the flowchart of the method of the present invention.

[0121] Figure 2 is the structure diagram of the system of the present invention.

[0122] Figure 3 It is a schematic diagram of the scheduling response of three flexible loads in the present invention.

[0123] Figure 4 It is a structural diagram of the device described in Embodiment 3. Detailed implementation manners

[0124] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0125] Embodiment 1:

[0126] Refer to Figure 1 and Figure 3 , a method for optimizing the photovoltaic carrying capacity based on multiple flexible loads, the optimization method includes:

[0127] S1. Construct an optimization model for photovoltaic carrying capacity. The optimization model for photovoltaic carrying capacity aims to minimize the multiple flexible loads. Refer to Table 1. The multiple flexible loads include transferable loads, shiftable loads, and curtailable loads, and consider the constraints affecting the flexible loads.

[0128] Table 1 Classification of flexible loads

[0129]

[0130] S2. Solve the optimization model for photovoltaic carrying capacity. Solve the optimization model for photovoltaic carrying capacity through an improved co-evolution algorithm to obtain an optimization plan for photovoltaic carrying capacity.

[0131] The transferable load is a load with a constant total power consumption within the scheduling period, but the power consumption in each period can be flexibly adjusted. The shiftable load is a load restricted by the production process and can only achieve a large-time shift of the power consumption curve. The curtailable load is a load whose power consumption can be reduced or interrupted through demand analysis.

[0132] The optimization model for photovoltaic carrying capacity includes an undispatchable load model and a dispatchable load model;

[0133] Both dispatchable flexible loads and undispatchable flexible loads have a certain degree of flexibility, that is, they can adjust their power consumption behaviors within a certain range. This flexibility provides more flexibility and choices for the scheduling and operation of the power system;

[0134] The main difference between dispatchable flexible loads and undispatchable flexible loads lies in their dispatchability. Dispatchable flexible loads can actively respond to grid dispatch instructions or price signals and optimize the operation of the power system by adjusting the power consumption time and amount, while undispatchable flexible loads have poor dispatchability due to various factors and are difficult to be dispatched in actual operation;

[0135] In the power system, there is a relationship of mutual influence and restriction between schedulable flexible loads and non-schedulable flexible loads. On the one hand, the scheduling behavior of schedulable flexible loads will be affected by non-schedulable flexible loads because the electricity consumption behavior of non-schedulable flexible loads will have a certain impact on the supply-demand balance of the power system. On the other hand, with the development of smart grids and power markets, through technical means and policy guidance, the schedulability of non-schedulable flexible loads can be gradually improved to better serve the operation of the power system.

[0136] Schedulable flexible loads are those that not only have time flexibility but can also actively participate in grid operation control and interact with the grid in terms of energy. Such loads can adjust their electricity consumption time and amount within a certain range to respond to grid scheduling instructions or electricity price signals. For example, electric vehicle charging, charging and discharging of energy storage devices, and some industrial and commercial loads, etc.; Non-schedulable flexible loads refer to those loads that, although having flexible characteristics (such as adjustability), cannot be scheduled or adjusted due to certain reasons (such as user preferences, contract constraints, etc.), and also generally refer to some loads that are theoretically adjustable but are difficult to schedule in actual operation due to technical, economic, or policy factors. Although such loads have a certain degree of flexibility, due to various factors, their schedulability is poor. For example, the air-conditioning loads of some residential users, although theoretically the electricity consumption can be reduced by adjusting the set temperature, it is actually very difficult to schedule due to the high demand for comfort by users.

[0137] The non-schedulable load model includes:

[0138] ΔD i trans = f1(D 0i ,Δp i ,ε ij ,v i trans );

[0139]

[0140] ΔD i shift = f2(i + Δi(Δp i )) - f2(i);

[0141] ΔD i re = f3(D 0i ,Δp i ,ε ii ,v i re );

[0142] In the above formula, ΔD itrans For i The response volume of the time-shiftable load, D 0i Is the reference load in the i-th period, △p i Is the electricity price difference vector between the i-th period and other periods, ε ij Is the cross-elasticity vector of the i-th period relative to the j-th period, v i trans Is the load transfer rate in the i-th period, ΔD i shift For i The response volume of the time-shiftable load, T is the scheduling period, ΔD i re Is the response volume of the load that can be curtailed in the i-th period, ε ii Is the self-elasticity vector, v i re Is the load curtailment rate;

[0143] The non-schedulable load model includes a load curtailment model, a time-shiftable load model, and a time-transferable load model. The load curtailment model includes:

[0144]

[0145] In the above formula, Is the power of the i-th load that can be curtailed in the t time period after participating in the demand response; μ i,t Indicates whether the i-th load that can be curtailed is curtailed in the t time period, μ i,t = 1 means the i-th load that can be curtailed is curtailed at time Segment; α i Is the curtailment rate of the i-th load that can be curtailed, 0 < α i < 1; Is the power of the i-th load that can be curtailed in the t time period before participating in the demand response; Is the capacity cost compensation for the i-th load that can be curtailed; c i Is the reported unit capacity cost of the i-th load that can be curtailed; Represents the operating cost of the i-th load that can be curtailed; Is the electricity price cost compensation for the i-th load that can be curtailed; T represents the scheduling period;

[0146] The time-shiftable load model includes:

[0147]

[0148] In the above formula, Is the operating cost of the i-th time-shiftable load; Is the cost per unit power of the i-th time-shiftable load; y i,τ Is the state of the i-th time-shiftable load, if yi,τ = 1 indicates that the load starts from period τ; is the acceptable shifting period range of the load; is the starting time period of the load; is the sum of the electrical powers of the i-th type of shiftable load; t Di is the total time period of electricity consumption of the i-th type of shiftable load. The total load includes is the set of starting time periods of the i-th type of shiftable load;

[0149] The said transferable load model includes:

[0150]

[0151] In the above formula, is the operating cost of the i-th type of transferable load; is the cost per unit power of the i-th type of transferable load; is the sum of the electrical powers of the i-th type of transferable load.

[0152] The said load curtailment constraint includes:

[0153] The curtailment duration constraint for the said load curtailment includes a minimum duration constraint and a maximum duration constraint. The minimum duration constraint includes:

[0154]

[0155] In the above formula, μ i,t indicates whether the i-th type of load curtailment is curtailed in the t time period. μ i,t = 1 means the i-th type of load curtailment is curtailed in the t time period. F i is the minimum curtailment time of the i-th type of load curtailment, is the minimum curtailment time of the i-th type of load curtailment, is the starting curtailment time of the i-th type of load curtailment;

[0156] The maximum duration constraint includes:

[0157]

[0158] In the above formula, G i is the maximum curtailment time of the i-th type of load curtailment;

[0159] The said shiftable load constraint includes:

[0160]

[0161] In the above formula, y i,τ indicates whether the i-th type of shiftable load starts from period τ, yi,τ = 1 indicates that the load starts from the τ period; is the set of start periods of the i-th type of shiftable load;

[0162] The shiftable load constraints include the minimum duration constraint, the load power constraint, the shiftable interval constraint, and the power balance constraint. The minimum duration constraint includes:

[0163]

[0164] In the above formula, represents the minimum duration of the i-th type of shiftable load, and v i,τ is whether the i-th type of shiftable load starts from the τ period, and v i,τ = 1 indicates that the load starts from the τ period; is the left limit point of the time interval of the i-th type of shiftable load, is the right limit point of the time interval of the i-th type of shiftable load.

[0165] The load power constraint includes:

[0166]

[0167] In the above formula, and represent the upper and lower limits of the power of the i-th type of shiftable load respectively, is the power of the i-th type of shiftable load at time t;

[0168] The power balance constraint includes:

[0169]

[0170] In the above formula, W i * represents the sum of the electricity quantities of the i-th type of shiftable load before being called, and Δt is the operating time of the shiftable load.

[0171] The S2 includes:

[0172] S2.1, according to the type of control quantity to be optimized, the control variables of the optimization model are divided into 3 populations: the load to be curtailed, the shiftable load, and the shiftable load. The operating cost component of the load to be curtailed X1 = [C1 re , C2 re , …, C i re ; The operating cost component of the shiftable load X2 = [C1 shift , C2 shift , …, C i shift ; The operating cost component of the shiftable load X3 = [C1trans , C2 trans , …, C i trans ;

[0173] S2.2, Generate the initial population individuals according to the improved initial population generation strategy, and the expression of the improved initial population generation strategy is:

[0174]

[0175] In the above formula, is the initial individual of the s-th population, s = 1, 2, …, S size ; r s is a random number uniformly distributed on [0, 1] for the s-th initial individual; A and B are the lower and upper limit value sets of the control variables respectively;

[0176] S2.3, Use the LHS method to sample the flexible load power / light intensity random variables to obtain D independent random samples;

[0177] S2.4, Perform probabilistic power flow calculation on the distribution network to test the feasibility of the individuals in the initial population, successively check whether the state variables of the D independent random samples exceed the limits, and judge whether the individuals meet the chance-constrained confidence level requirements according to the law of large numbers. If not, perform contraction and approach according to the following formula until they become feasible individuals:

[0178]

[0179] In the above formula, α is the contraction coefficient, with an initial value of 0.5, and the value of α is halved in each contraction process, and finally will become a feasible initial individual;

[0180] S2.5, Repeat S2.2 to S2.4 until S size initial populations that meet the chance constraints are generated;

[0181] S2.6, Perform selection, crossover, and mutation operations on each sub-population according to the genetic algorithm process to generate new individuals Px', and the new individual Px' is the node power component of a group of flexible loads with the largest fitness function value;

[0182] S2.7, Coordinate between populations, select the individual with the highest fitness value from the other population, and after decoding, jointly form the control variable Px” of the system with the individual Px' within its own population, and calculate the fitness of the individual Px';

[0183] S2.8, Judge whether the termination condition for overall collaborative optimization is met, and the termination condition is to reach the maximum genetic algebra k max, when the termination condition is not met, enter S2.6 to perform re-optimization calculation. When the termination condition is met, terminate the optimization process to obtain the Pareto optimal solution set and enter S2.9;

[0184] S2.9, use the fuzzy method based on information entropy to select the optimal compromise solution from the obtained Pareto optimal solution set, and finally obtain a set of optimal multi-flexible load combinations to achieve the improvement of the photovoltaic carrying capacity.

[0185] Embodiment 2:

[0186] See Figure 2 , a photovoltaic carrying capacity optimization system based on multi-flexible loads, the system includes a model construction module and a model solution module;

[0187] The model construction module is used to construct a photovoltaic carrying capacity optimization model, which aims to minimize the multi-flexible load. The multi-flexible load includes shiftable load, shiftable load and curtailable load, and considers the constraints affecting the flexible load;

[0188] The model solution module is used to solve the photovoltaic carrying capacity optimization model by an improved co-evolution algorithm to obtain a photovoltaic carrying capacity optimization plan.

[0189] The shiftable load is a load with a constant total power consumption within the scheduling period, but the power consumption in each period can be flexibly adjusted. The shiftable load is a load that is restricted by the production process and can only achieve a large-time shift of the power consumption curve. The curtailable load is a load whose power consumption can be reduced or interrupted through demand analysis.

[0190] The photovoltaic carrying capacity optimization model includes an unschedulable load model and a schedulable load model. The unschedulable load model includes:

[0191] ΔD i trans = f1(D 0i ,Δp i ,ε ij ,v i trans );

[0192]

[0193] ΔD i shift = f2(i + Δi(Δp i )) - f2(i);

[0194] ΔD i re = f3(D 0i ,Δp i ,εii , v i re );

[0195] In the above formula, ΔD i trans is i the response volume of the shiftable load in the time period, D 0i is the reference load in the i-th time period, △p i is the electricity price difference vector between the i-th time period and other time periods, ε ij is the cross-elasticity vector of the i-th time period relative to the j-th time period, v i trans is the load transfer rate in the i-th time period, ΔD i shift is i the response volume of the shiftable load in the time period, T is the scheduling period, ΔD i re is the response volume of the load that can be curtailed in the i-th time period, ε ii is the self-elasticity vector, v i re is the load curtailment rate;

[0196] The non-schedulable load model includes a load curtailment model, a shiftable load model, and a shiftable load model. The load curtailment model includes:

[0197]

[0198] In the above formula, is the power of the i-th type of load that can be curtailed in the t time period after participating in the demand response; μ i,t indicates whether the i-th type of load that can be curtailed is curtailed in the t time period. μ i,t = 1 means that the i-th type of load that can be curtailed is curtailed in the t time period; α i is the curtailment rate of the i-th type of load that can be curtailed, 0 < α i < 1; is the power of the i-th type of load that can be curtailed in the t time period before participating in the demand response; is the capacity cost compensation of the i-th type of load that can be curtailed; c i is the reported unit capacity cost of the i-th type of load that can be curtailed; represents the operating cost of the i-th type of load that can be curtailed; is the electricity price cost compensation of the i-th type of load that can be curtailed; T represents the scheduling period;

[0199] The shiftable load model includes:

[0200]

[0201] In the above formula, is the operating cost of the i-th shiftable load; is the cost per unit power of the i-th shiftable load; y i,τ is the status of the i-th shiftable load. If y i,τ = 1, it means the load starts from the τ period; is the interval of shiftable periods acceptable to the load; is the starting time period of the load; is the sum of the power consumptions of the i-th shiftable load; t Di is the total time period of power consumption of the i-th shiftable load. The total load includes is the set of starting periods of the i-th shiftable load;

[0202] The said shiftable load model includes:

[0203]

[0204] In the above formula, is the operating cost of the i-th shiftable load; is the cost per unit power of the i-th shiftable load; is the sum of the power consumptions of the i-th shiftable load.

[0205] The said load curtailment constraint includes:

[0206] The curtailment duration constraint for the load curtailment includes a minimum duration constraint and a maximum duration constraint. The minimum duration constraint includes:

[0207]

[0208]

[0209] In the above formula, μ i,t indicates whether the i-th load curtailment load is curtailed in the t period. μ i,t = 1 means the i-th load curtailment load is curtailed in the t period. F i is the minimum curtailment time of the i-th load curtailment load, is the minimum curtailment time of the i-th load curtailment load, is the starting curtailment time of the i-th load curtailment load;

[0210] The said maximum duration constraint includes:

[0211]

[0212] In the above formula, G i is the maximum curtailment time of the i-th load curtailment load;

[0213] The translatable load constraint includes:

[0214]

[0215] In the above formula, y i,τ indicates whether the i-th translatable load starts from the τ period. y i,τ = 1 means the load starts from the τ period; is the set of start periods of the i-th translatable load;

[0216] The transferable load constraint includes a minimum duration constraint, a load power constraint, a transferable interval constraint, and an electric energy balance constraint. The minimum duration constraint includes:

[0217]

[0218] In the above formula, represents the minimum duration of the i-th transferable load, and v i,τ indicates whether the i-th transferable load starts from the τ period. v i,τ = 1 means the load starts from the τ period; is the left limit point of the time interval of the i-th transferable load, is the right limit point of the time interval of the i-th transferable load.

[0219] The load power constraint includes:

[0220]

[0221] In the above formula, and respectively represent the upper and lower limits of the power of the i-th transferable load, is the power of the i-th transferable load at the t time period;

[0222] The electric energy balance constraint includes:

[0223]

[0224] In the above formula, W i * represents the sum of the electric energy of the i-th transferable load before being called, and Δt is the operation time of the transferable load.

[0225] The specific operation steps of the model solving module are as follows:

[0226] S2.1, divide the optimization model control variables into three populations: curtailable load, translatable load, and transferable load according to the type of control quantity to be optimized. The operating cost component of the curtailable load X1 = [C1 re , C2 re , …, Ci re ; The translatable load operation cost component X2 = [C1 shift , C2 shift , …, C i shift ; The transferable load operation cost component X3 = [C1 trans , C2 trans , …, C i trans ;

[0227] S2.2. Generate initial population individuals according to the improved initial population generation strategy. The expression of the improved initial population generation strategy is:

[0228]

[0229] In the above formula, is the initial individual of the s-th population, s = 1, 2, …, S size ; r s is a random number uniformly distributed on [0, 1] for the s-th initial individual; A and B are the lower and upper limit value sets of the control variables respectively;

[0230] S2.3. Use the LHS method to sample the flexible load power / light intensity random variables to obtain D independent random samples;

[0231] S2.4. Perform probabilistic power flow calculation on the distribution network to test the feasibility of the individuals in the initial population. Successively check whether the state variables of the D independent random samples exceed the limits, and judge whether the individuals meet the opportunity constraint confidence level requirements according to the law of large numbers. If not, perform contraction and convergence according to the following formula until they become feasible individuals:

[0232]

[0233] In the above formula, α is the contraction coefficient, with an initial value of 0.5, and the value of α is halved in each contraction process, and finally will always become a feasible initial individual;

[0234] S2.5. Repeat S2.2 to S2.4 until S size initial populations that meet the opportunity constraints are generated;

[0235] S2.6. Perform selection, crossover, and mutation operations on each sub-population according to the genetic algorithm process to generate new individuals Px'. The new individual Px' is the node power component of a group of flexible loads with the largest fitness function value;

[0236] S2.7, Inter-population cooperation: Select the individual with the highest fitness value from the other population. After decoding, it jointly constitutes the control variable Px” of the system with the individual Px' within its own population, and calculate the fitness of the individual Px'.

[0237] S2.8, Determine whether the termination condition for overall cooperative optimization is satisfied. The termination condition is reaching the maximum genetic algebra k max , When the termination condition is not satisfied, enter S2.6 for re-optimization calculation. When the termination condition is satisfied, terminate the optimization process to obtain the Pareto optimal solution set and enter S2.9;

[0238] S2.9, Use the fuzzy method based on information entropy to select the optimal compromise solution from the obtained Pareto optimal solution set, and finally obtain a set of optimal multi-flexible load combinations to achieve the improvement of the photovoltaic carrying capacity.

[0239] Embodiment 3:

[0240] See Figure 4 , A photovoltaic carrying capacity optimization device based on multi-flexible load, the device includes a processor and a memory;

[0241] The memory is used to store computer program code and transmit the computer program code to the processor;

[0242] The processor is used to execute the photovoltaic carrying capacity optimization method based on multi-flexible load described in Embodiment 1 according to the instructions in the computer program code.

[0243] A computer medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the photovoltaic carrying capacity optimization method based on multi-flexible load described in Embodiment 1.

[0244] The above are only the preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. Any equivalent modifications or changes made by those of ordinary skill in the art according to the content disclosed in the present invention shall be included in the protection scope recorded in the claims.

Claims

1. A photovoltaic carrying capacity optimization method based on multi-element flexible loads, characterized in that: The optimization method comprises: S1. Constructing a photovoltaic carrying capacity optimization model. Constructing a photovoltaic carrying capacity optimization model. The photovoltaic carrying capacity optimization model aims to minimize the multivariate flexible load. The multivariate flexible load includes a transferable load, a shiftable load, and a reducible load, and considers the constraints that affect the flexible load; S2. Solve the photovoltaic carrying capacity optimization model. Solve the photovoltaic carrying capacity optimization model by improving the collaborative evolution algorithm to obtain the photovoltaic carrying capacity optimization plan.

2. The photovoltaic carrying capacity optimization method based on multi-element flexible load according to claim 1, characterized in that: The transferable load refers to a load whose total power consumption remains unchanged during the scheduling cycle, but whose power consumption in each time period can be flexibly adjusted. The shiftable load refers to a load that is constrained by the production process and can only shift the power consumption curve over a large time period. The reducible load refers to a load whose power consumption can be reduced or interrupted through demand analysis.

3. The photovoltaic carrying capacity optimization method based on multi-element flexible load according to claim 2, characterized in that: The photovoltaic carrying capacity optimization model includes a non-dispatchable load model and a dispatchable load model, and the non-dispatchable load model includes: ΔD i trans =f1(D 0i ,Δp i ,he ij ,v i trans ); ΔD i shift =f2(i+Δi(Δp i ))-f2(i); ΔD i re =f3(D 0i ,Δp i ,he ii ,v i re ); In the above formula, ΔD i trans is the response amount of the load that can be transferred in period i, D 0i is the benchmark load in period i, △p i is the price difference vector between period i and other periods, ε ij is the mutual elasticity vector of period i relative to period j, v i trans for i Load transfer rate during the period, ΔD i shift for i The response amount of the load that can be shifted during the time period, T is the scheduling period, ΔD i re is the load reduction response in period i, ε ii is the self-elastic vector, v i re is the load reduction rate; The non-dispatchable load model includes a curtailable load model, a shiftable load model and a transferable load model. The curtailable load model includes: In the above formula, is the power of the i-th load that can be reduced in time period t after participating in demand response; μ i,t Indicates whether the i-th type of curtailable load is curtailed in time period t, μ i,t =1 is the i-th type of load that can be reduced at time t The segment is cut; α i is the reduction rate of the i-th type of reducible load, is the power of the i-th type of curtailable load in time period t before participating in demand response; Compensation for the i-th type of load capacity that can be reduced; c i The unit capacity fee reported for the i-th curtailable load; represents the operating cost of the i-th type of load that can be reduced; is the electricity price compensation for the i-th type of reducible load; T represents the dispatch period; The translatable load model includes: In the above formula, is the operating cost of the i-th type of movable load; is the cost per unit power of the i-th type of movable load; y i,τ is the state of the i-th translatable load, if y i,τ =1 means the load starts from the τ period; The load translation time interval is acceptable; is the starting time period of the load; is the sum of the power consumption of the ith type of movable load; t Di is the total power consumption period of the i-th type of shiftable load, and the total load includes is the starting time set of the i-th type of translatable load; The transferable load model includes: In the above formula, is the operating cost of the i-th transferable load; is the cost per unit power of the i-th transferable load; is the sum of the power consumed by the i-th transferable load.

4. The photovoltaic carrying capacity optimization method based on multi-element flexible load according to claim 3, characterized in that: The curtailable load constraints include: The load shedding duration constraint that constrains the load shedding includes a minimum duration constraint and a maximum duration constraint, wherein the minimum duration constraint includes: In the above formula, μ i,t Indicates whether the i-th type of curtailable load is curtailed in time period t, μ i,t =1 means the i-th type of curtailable load is curtailed in time period t, F i is the minimum reduction time of the i-th type of load that can be reduced, is the minimum reduction time of the i-th type of load that can be reduced, is the starting reduction time of the i-th type of load that can be reduced; The maximum duration constraints include: In the above formula, G i is the maximum reduction time of the i-th type of load that can be reduced; The translatable load constraints include: In the above formula, y i,τ is whether the i-th type of translatable load starts from the τ period, y i,τ =1 means the load starts from the τ period; is the starting time set of the i-th type of translatable load; The transferable load constraints include a minimum duration constraint, a load power constraint, a transferable interval constraint, and an electric energy balance constraint. The minimum duration constraint includes: In the above formula, represents the minimum duration of the i-th transferable load, v i,τ is whether the i-th transferable load starts from the τ period, v i,τ =1 means the load starts from the τ period; is the left limit point of the i-th transferable load time interval, is the right limit point of the i-th transferable load time interval. The load power constraints include: In the above formula, and Respectively represent the upper and lower limits of the power of the i-th transferable load, is the power of the i-th transferable load in time period t; The power balance constraints include: In the above formula, W i * It represents the sum of the electricity of the i-th transferable load before it is called, and Δt is the operating time of the transferable load.

5. The photovoltaic carrying capacity optimization method based on multi-element flexible load according to claim 4, characterized in that: The S2 includes: S2.1, according to the type of control quantity to be optimized, the control variables of the optimization model are divided into three groups: reducible load, shiftable load and transferable load. The reducible load operation cost component X1 = [C1 re , C2 re , …, C i re ]; the movable load operating cost component X2 = [C1 shift , C2 shift , …, C i shift ]; the transferable load operating cost component X3 = [C1 trans , C2 trans , …, C i trans ]; S2.2, generating initial population individuals according to the improved initial population generation strategy, the improved initial population generation strategy expression is: In the above formula, is the initial individual of the sth population, s = 1, 2, ..., S size ; r s is a random number of the sth initial individual that follows a uniform distribution on [0, 1]; A and B are the lower and upper limit value sets of the control variable respectively; S2.3, use the LHS method to sample the flexible load power / light intensity random variable to obtain D independent random samples; S2.4, perform probabilistic power flow calculation on the distribution network, check the feasibility of individuals in the initial population, check whether the state variables of D independent random samples exceed the limit, and judge whether the individual meets the confidence requirement of the chance constraint according to the law of large numbers. If not, shrink and converge according to the following formula until it becomes a feasible individual: In the above formula, α is the contraction coefficient, the initial value is 0.5, and the value of α is halved in each contraction process, and the final result is Become a viable initial individual; S2.5, repeat S2.2 to S2.4 until S is generated size An initial population that satisfies the chance constraint; S2.6, performing selection, crossover and mutation operations on each subpopulation according to the genetic algorithm process to generate a new individual Px', wherein the new individual Px' is a node power component of a group of flexible loads with the largest fitness function value; S2.7, inter-population collaboration, select the individual with the highest fitness value from the other population, and after decoding, together with the individual Px' in its own population, form the control variable Px' of the system, and calculate the fitness of the individual Px'; S2.8, determine whether the termination condition of the overall collaborative optimization is met, the termination condition is to reach the maximum genetic generation number k max , when the termination condition is not met, enter S2.6 and re-calculate the optimization. When the termination condition is met, terminate the optimization process, obtain the Pareto optimal solution set, and enter S2.9; S2.9, using the fuzzy method based on information entropy, the optimal compromise solution is selected from the Pareto optimal solution set, and finally a set of optimal multivariate flexible load combinations is obtained to achieve the improvement of photovoltaic carrying capacity.

6. A photovoltaic carrying capacity optimization system based on multi-element flexible loads, characterized in that: The system includes a model building module and a model solving module; The model building module is used to build a photovoltaic carrying capacity optimization model, wherein the photovoltaic carrying capacity optimization model aims to minimize the multivariate flexible load, wherein the multivariate flexible load includes a transferable load, a translationally movable load, and a reducible load, and considers the constraints affecting the flexible load; The model solving module is used to solve the photovoltaic carrying capacity optimization model through an improved collaborative evolution algorithm to obtain a photovoltaic carrying capacity optimization solution.

7. The photovoltaic carrying capacity optimization system based on multi-element flexible loads according to claim 6, characterized in that: The transferable load refers to a load whose total power consumption remains unchanged during the scheduling cycle, but whose power consumption in each time period can be flexibly adjusted. The shiftable load refers to a load that is constrained by the production process and can only shift the power consumption curve over a large time period. The reducible load refers to a load whose power consumption can be reduced or interrupted through demand analysis.

8. The photovoltaic carrying capacity optimization system based on multi-element flexible loads according to claim 7, characterized in that: The photovoltaic carrying capacity optimization model includes a non-dispatchable load model and a dispatchable load model, and the non-dispatchable load model includes: ΔD i trans =f1(D 0i ,Δp i ,he ij ,v i trans ); ΔD i shift =f2(i+Δi(Δp i ))-f2(i); ΔD i re =f3(D 0i ,Δp i ,he ii ,v i re ); In the above formula, ΔD i trans for i The response amount of the load that can be transferred during the period, D 0i is the benchmark load in period i, △p i is the price difference vector between period i and other periods, ε ij is the mutual elasticity vector of period i relative to period j, v i trans is the load transfer rate in period i, ΔD i shift for i The response amount of the load that can be shifted during the time period, T is the scheduling period, ΔD i re is the load reduction response in period i, ε ii is the self-elastic vector, v i re is the load reduction rate; The non-dispatchable load model includes a curtailable load model, a shiftable load model and a transferable load model. The curtailable load model includes: In the above formula, is the power of the i-th load that can be reduced in time period t after participating in demand response; μ i,t Indicates whether the i-th type of curtailable load is curtailed in time period t, μ i,t =1 is the i-th type of load that can be reduced at time t The segment is cut; α i is the reduction rate of the i-th type of load that can be reduced, 0<α i <1; is the power of the i-th type of curtailable load in time period t before participating in demand response; Compensation for the i-th type of load capacity that can be reduced; c i The unit capacity fee reported for the i-th curtailable load; represents the operating cost of the i-th type of load that can be reduced; is the electricity price compensation for the i-th type of reducible load; T represents the dispatch period; The translatable load model includes: In the above formula, is the operating cost of the i-th type of movable load; is the cost per unit power of the i-th type of movable load; y i,τ is the state of the i-th translatable load, if y i,τ =1 means the load starts from the τ period; The load translation time interval is acceptable; is the starting time period of the load; is the sum of the power consumption of the ith type of movable load; t Di is the total power consumption period of the i-th type of shiftable load, and the total load includes is the starting time set of the i-th type of translatable load; The transferable load model includes: In the above formula, is the operating cost of the i-th transferable load; is the cost per unit power of the i-th transferable load; is the sum of the power consumed by the i-th transferable load.

9. The photovoltaic carrying capacity optimization system based on multi-element flexible loads according to claim 8, characterized in that: The curtailable load constraints include: The load shedding duration constraint that constrains the load shedding includes a minimum duration constraint and a maximum duration constraint, wherein the minimum duration constraint includes: In the above formula, μ i,t Indicates whether the i-th type of curtailable load is curtailed in time period t, μ i,t =1 means the i-th type of curtailable load is curtailed in time period t, F i is the minimum reduction time of the i-th type of load that can be reduced, is the minimum reduction time of the i-th type of load that can be reduced, is the starting reduction time of the i-th type of load that can be reduced; The maximum duration constraints include: In the above formula, G i is the maximum reduction time of the i-th type of load that can be reduced; The translatable load constraints include: In the above formula, y i,τ is whether the i-th type of translatable load starts from the τ period, y i,τ =1 means the load starts from the τ period; is the starting time set of the i-th type of translatable load; The transferable load constraints include a minimum duration constraint, a load power constraint, a transferable interval constraint, and an electric energy balance constraint. The minimum duration constraint includes: In the above formula, represents the minimum duration of the i-th transferable load, v i,τ is whether the i-th transferable load starts from the τ period, v i,τ =1 means the load starts from the τ period; is the left limit point of the i-th transferable load time interval, is the right limit point of the i-th transferable load time interval. The load power constraints include: In the above formula, and Respectively represent the upper and lower limits of the power of the i-th transferable load, is the power of the i-th transferable load in time period t; The power balance constraints include: In the above formula, W i * It represents the sum of the electricity of the i-th transferable load before it is called, and Δt is the operating time of the transferable load.

10. A photovoltaic carrying capacity optimization system based on multi-element flexible loads according to claim 9, characterized in that: The specific operation steps of the model solving module are: S2.1, according to the type of control quantity to be optimized, the control variables of the optimization model are divided into three groups: reducible load, shiftable load and transferable load. The reducible load operation cost component X1 = [C1 re , C2 re , …, C i re ]; the movable load operating cost component X2 = [C1 shift , C2 shift , …, C i shift ]; the transferable load operating cost component X3 = [C1 trans , C2 trans , …, C i trans ]; S2.2, generating initial population individuals according to the improved initial population generation strategy, the improved initial population generation strategy expression is: In the above formula, is the initial individual of the sth population, s = 1, 2, ..., S size ; r s is a random number of the sth initial individual that follows a uniform distribution on [0, 1]; A and B are the lower and upper limit value sets of the control variable respectively; S2.3, use the LHS method to sample the flexible load power / light intensity random variable to obtain D independent random samples; S2.4, perform probabilistic power flow calculation on the distribution network, check the feasibility of individuals in the initial population, check whether the state variables of D independent random samples exceed the limit, and judge whether the individual meets the confidence requirement of the chance constraint according to the law of large numbers. If not, shrink and converge according to the following formula until it becomes a feasible individual: In the above formula, α is the contraction coefficient, the initial value is 0.5, and the value of α is halved in each contraction process, and the final result is Become a viable initial individual; S2.5, repeat S2.2 to S2.4 until S is generated size An initial population that satisfies the chance constraint; S2.6, performing selection, crossover and mutation operations on each subpopulation according to the genetic algorithm process to generate a new individual Px', wherein the new individual Px' is a node power component of a group of flexible loads with the largest fitness function value; S2.7, inter-population collaboration, select the individual with the highest fitness value from the other population, and after decoding, together with the individual Px' in its own population, form the control variable Px' of the system, and calculate the fitness of the individual Px'; S2.8, determine whether the termination condition of the overall collaborative optimization is met, the termination condition is to reach the maximum genetic generation number k max , when the termination condition is not met, enter S2.6 and re-calculate the optimization. When the termination condition is met, terminate the optimization process, obtain the Pareto optimal solution set, and enter S2.9; S2.9, using the fuzzy method based on information entropy, the optimal compromise solution is selected from the Pareto optimal solution set, and finally a set of optimal multivariate flexible load combinations is obtained to achieve the improvement of photovoltaic carrying capacity.

Images