CN105279578A - Power supply optimization configuration bilevel programming method in active distribution network region - Google Patents

Abstract

The present invention provides a power supply optimization configuration bilevel programming method in an active distribution network region. The method comprises the following steps: basic data acquisition, model construction, model transformation, algorithm parameter initialization, initial network loss resolution, power generation cost correction per unit power, upper planning scheme resolution, lower planning scheme resolution, iteration termination condition discrimination and the like. Based on an integrated resource strategy planning theory, the method provided by the present invention considers a generalized power supply represented by a photovoltaic power source and an interruptible load from a whole society angle, and establishes the power supply extended optimization bilevel programming model in the active distribution network region. The method comprises: according to the upper-level planning scheme, for the purpose of the lowest total cost of general power supply construction, power generation and pollution treatment, optimizing each power supply installed capacity; according to the lower-level planning scheme, for the purpose of minimum electrical power loss, locating and sizing the general power supply; and finally performing resolution in combination with a simplex method and an improved PSO algorithm. Compared with the conventional programming method, the bilevel programming method provided by the present invention has a significant effect of energy saving and emission reduction, a high resource utilization rate and an improved whole social benefit.

Description

A two-level programming method for optimal allocation of regional power sources in active distribution networks

technical field

The invention belongs to the technical field of power distribution system planning, and in particular relates to a two-layer planning method for optimal configuration of regional power sources in an active distribution network.

Background technique

With the shortage of traditional fossil power sources and environmental pollution becoming more and more prominent, the traditional power grid planning method that relies solely on increasing power sources and investment in substations to meet the rapidly growing power demand has been greatly challenged by various aspects. The rapid development of distributed generation technology (DistributedGeneration, referred to as DG) and demand response technology provides ideas for solving these problems. However, the traditional distribution network is not capable of accommodating renewable power sources, so it cannot meet the requirements of large-scale access to new power sources. At the same time, its planning method does not consider the impact of DG introduction on the distribution network, and the planning scheme is too conservative, so it cannot Make full use of grid assets.

In recent years, the active distribution network (Active Distribution Network, referred to as ADN), which has been hotly discussed in academia and industry, has the ability to combine and control various distributed power sources (DG, controllable load, energy storage, demand-side management, etc.), which can enhance power distribution. The ability of the network to accept distributed power, realize the interaction between supply and demand, and reduce pollutant emissions and waste of resources. At the same time, ADN helps to improve the utilization rate of distribution network assets, delay equipment upgrade investment, and improve users' power quality and power supply reliability. In order to be actively compatible with renewable power sources and demand-side resources, ADN planning needs to comprehensively consider multiple factors such as economics, environmental benefits, and resource utilization, and because its investment involves power companies, distributed power providers, and demand cluster response resource suppliers, etc. This will inevitably promote the transformation of ADN planning from the traditional pursuit of maximizing the interests of a single subject to the direction of complex multi-subject coordinated planning.

Introduce the theory of Integrated Resource Strategy Planning (IRSP) to integrate and optimize power supply-side resources and various forms of power demand-side resources, and rationally and effectively integrate supply from a strategic perspective through economic, legal and administrative means On the premise of meeting the power demand of future economic development, the resources on the side and the demand side can reduce the input cost of the whole society, the consumption of power resources and the discharge of pollutants, and provide power users with the lowest cost and the most comprehensive benefits of power services, becoming ADN The inevitable requirement of comprehensive planning.

Over the years, there have been rich research results accumulated in the traditional distribution network planning at home and abroad, and some progress has been made in the research of active distribution network planning in recent years. However, most studies only build planning models from the perspective of power company investment and profitability. From the perspective of the whole society, the research on active distribution network planning that incorporates supply-side and demand-side resources into planning at the same time remains to be in-depth.

Contents of the invention

In order to solve the above problems, the object of the present invention is to provide a bi-level programming method for optimal allocation of regional power sources in an active distribution network.

In order to achieve the above purpose, (I will copy it after the content of the claim is determined here)

Based on the theory of comprehensive resource strategic planning, the present invention considers the generalized power source represented by photovoltaic power source and interruptible load from the perspective of the whole society, and establishes a two-layer programming model for the expansion and optimization of regional power source of active distribution network. The upper-level planning model aims to minimize the total cost of generalized power supply construction, power generation and pollution control, and optimizes the installed capacity of various power sources; the lower-level planning model aims to minimize network loss, and selects the location and capacity of the generalized power supply. Finally, the simplex method and the improved PSO algorithm are combined to solve the problem. Compared with the traditional planning method, the planning method of the present invention has obvious energy saving and emission reduction effects, higher resource utilization rate, and more prominent social benefits, and can provide guidance for the planning, construction and operation of the active distribution network.

The beneficial effects of the two-layer planning method for optimal allocation of regional power sources in the active distribution network provided by the present invention:

(1) The model description is more comprehensive: When solving the active distribution network expansion and distributed power access planning problems, the IRSP two-tier programming model can plan the planning capacity of various power sources taking into account the total cost of the whole society, and simulate the distribution network According to the actual operation situation, the active distribution network configuration scheme with the smallest network loss of the system is obtained by solving. At the same time, in the optimization iteration, using the relationship between system operation network loss and power generation cost to correct the power planning results can ensure the highest level of comprehensive utilization of the overall power resources and minimize the cost of the whole society.

(2) Environmental benefits are more prominent: Using IRSP theory to guide power grid planning can take into account environmental impacts, taking into account the energy saving and emission reduction benefits of distributed power sources and demand-side management methods, and is more suitable for the development of building an environment-friendly society in the future than traditional planning theories trend.

Description of drawings

Fig. 1 is the planning diagram of the amount of power supplied by various power sources in the middle-aged continuous load curve of the present invention;

Fig. 2 is a flow chart of a two-layer planning method for optimal configuration of active distribution network regional power sources provided by the present invention;

Figure 3 is the topological diagram of the network frame structure and the node number diagram of the 10kV33 node calculation example;

detailed description

The bi-level programming method for optimal allocation of regional power sources in active distribution networks provided by the present invention will be described in detail below in conjunction with the drawings and embodiments.

Taking the 33-node calculation example shown in FIG. 3 as an example, the two-layer planning method for optimal allocation of regional power sources in the active distribution network provided by the present invention will be described in detail in combination with the flow chart shown in FIG. 2 .

As shown in Figure 2, the bilevel programming method for optimal configuration of regional power sources in active distribution networks provided by the present invention includes the following steps executed in sequence:

Step 1. Acquisition of basic data: Obtain the basic data of the power distribution system to be studied, including power supply type, grid structure, load level, and electrical parameters;

In this embodiment, basic data such as power source type, grid structure, load level, and electrical parameters of the 33-node power distribution system are obtained. Among them, it is assumed that the load level of the system in the planning year is 1.7 times of the original, and the basic load is supplied by the original power supply of the system, and the excess part is supplied by the original power supply of the system and the new power supply. Photovoltaic power sources and signed interruptible load contracts represent three different types of power sources: power substations, distributed power sources, and demand-side resources in the distribution network. The grid structure is shown in Figure 3, and the node numbers are shown in the figure. The parameters of the load level are shown in Table 1, and the electrical parameters of the branches are shown in Table 2. The candidate installation nodes of the photovoltaic power supply are set to 7, 11, 15, 18, 29, 32, and the candidate installation nodes of the interruptible load are 8, 14, 21, 24, 30, the interrupted load is interrupted according to the original power factor;

Table 133 Node calculation example load level parameters

Table 2 Branch Electrical Parameters

Step 2. Model construction: use the basic data obtained in step 1 to construct a two-level programming model for the regional power expansion problem based on IRSP, and determine the upper and lower objective functions and constraints of the two-level programming model;

In step 2, the two-level programming model of the regional power supply expansion problem based on IRSP includes upper and lower planning models. The cost is the lowest, and its objective function includes the initial investment construction cost, power generation cost and pollutant treatment cost of various power sources. It is a minimization problem, and its mathematical expression is:

$\mathrm{<span\; class="notranslate">\; min</span>}\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; I</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; I</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

In the formula, i is the type of power supply, i=1, 2, and 3 respectively represent thermal power generation, photovoltaic power supply, and interruptible load; F _i is the annual value of the construction cost per unit capacity of this type of power supply, yuan/kW; C _i is the type of power supply Total installed capacity, kW; V _i is the power generation cost per unit of electricity of this type of power supply, yuan/kWh; H _i is the annual utilization hours of this type of power supply, hours; β _i is the pollutant treatment cost per unit of electricity of this type of power supply, yuan/kWh kWh. F _i can be further expressed as:

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&CenterDot;</span>\frac{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; m</span>}}}{{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

In the formula, a _i is the unit capacity cost of this type of power supply, yuan/kW; r ₀ is the discount rate; m is the operating life of this type of power supply (calculated life span years), years.

The annual operating hours of photovoltaic power sources and interruptible loads are limited by sunshine factors and contract content, making it difficult to provide long-term power supply. The lack of power supply is made up by thermal power generation. In order to avoid the planned capacity of photovoltaic power supply and interruptible load being too large, and the planned capacity of thermal power generation too small, resulting in a shortage of actual power supply, the upper-level planning model should satisfy the constraints that the total power supply of all power sources is greater than or equal to the annual power demand of the planning area. :

$\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; I</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>{<span\; class="notranslate">\; \&Integral;</span>}_{\mathrm{<span\; class="notranslate">\; 0</span>}}^{\mathrm{<span\; class="notranslate">\; 8760</span>}}\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

In the formula, L(t) is the average load power per hour throughout the year.

The lower-level planning model is used to realize the optimization of various power supply access point selection and system operation simulation optimization at the micro-level operation level, so as to ensure the minimum network loss during system operation. The objective function of the lower-level planning model should consider the relationship between the operating costs of various power sources and the peak-shaving characteristics. When calculating the total annual operating costs, the total capacity of distributed power sources and interruptible loads given by the upper-level planning of the IRSP should be allocated to each power source. Select the access point, through the simulation of the actual operation status of the power grid for 24 hours in a typical day, with the goal of minimizing the power loss of the network, optimize the allocation capacity of the access point to be selected. The mathematical expression is:

minf=loss(4)

The constraints are:

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; I</span>}\mathrm{<span\; class="notranslate">\; L</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; cos\&delta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; sin\&delta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; I</span>}\mathrm{<span\; class="notranslate">\; L</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; sin\&delta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; cos\&delta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

U _imin ≤ U _i ≤ U _imax (7)

_{0≤Iw≤Ijmax (} 8)

S _j ≤ S j _max (9)

In the formula, loss is the power loss of the network; n is the number of system nodes; P _pv , Q _pv , P _IL , Q _IL are the active and reactive power injected into node i by distributed photovoltaic and interruptible load respectively; P _i , Q _i are the active and reactive loads of node i respectively; G _ij , B _ij are the corresponding elements in the node admittance matrix; U _i is the voltage amplitude of node i, U _imin , U _imax are the allowable voltage of node i Upper limit and lower limit; I _j is the current amplitude of branch j, I _jmax is the upper limit of current thermal stability of branch j; S _j is the apparent power of branch j, and S _jmax is the upper limit of apparent power of branch j.

In this embodiment, the upper and lower limits of the system node voltage are 1.05p.u. and 0.95p.u., the maximum current allowed to flow through the branch is 0.4kA, and the upper limit of the branch power is 6.93kVA.

Step 3. Algorithm parameter initialization: Initialize the parameters of the bi-level programming algorithm, set the maximum number of iterations of the improved PSO algorithm to 50, the number of particle populations to n, and the particle code length to be the photovoltaic power source and The total number of nodes that can interrupt the load, learning factors C1 = 2, C2 = 1.732, and determine the iteration termination condition;

In this embodiment, the number of particle populations is 30, the particle code length is 11 (6 nodes for photovoltaic power supply, and 5 nodes for interruptible load), learning factors C1=2, C2=1.732, determine A maximum of 50 iterations was reached as the termination condition.

Step 4. Initial network loss solution: Assuming that the capacity of the upper-level substation is sufficient, and the photovoltaic power source and interruptible load are not considered, the lower-level planning model is used to calculate the power loss of the network when all the loads of the active distribution network are supplied by the upper-level substation. The substation is responsible for thermal power generation;

Step 5. Correction of power generation cost per unit of electricity: apportion the network loss power obtained by solving the above-mentioned lower-level planning model to various power sources to correct the power generation costs of various power sources;

In Step 5, the power loss caused by power generation and transmission should be included in the cost of power generation, so the power generation cost V i of the power unit of type _i should be:

${\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; load</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; e</span>}\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; loss</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; load</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; e</span>}\mathrm{<span\; class="notranslate">\; m</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; )</span>$

In the formula, S _pdi is the unit power generation cost of the original i-type power source; load _dem is the load power consumption borne by the i-type power source; loss _i is the network loss power borne by the i-type power source. It can be seen that the unit power generation cost V _i of the i-type power supply is the dependent variable of load power consumption and network loss power.

The allocation of network loss and electricity of the active distribution network including distributed power is divided into two steps. The first step is to allocate the network loss and electricity of the active distribution network without distributed power to the original power supplier; the second step is to The change in network loss and electricity caused by the access of distributed power sources is apportioned to distributed power sources.

Δloss＝loss ₁ -loss′ ₁ (11)

loss ₂ = Δloss(12)

In the formula, Δloss is the difference between loss ₁ and loss ₁ ', loss ₁ is the total network loss after accessing distributed power, loss ₁ ' is the network loss without distributed power; loss ₂ is distributed Network power loss shared by the power supply.

After accessing the distributed power generation, if the network loss of the active distribution network decreases, Δloss is negative, that is, the network loss loss ₂ shared by the distributed power generation is a negative value, and the unit power generation cost of the photovoltaic power source can be known from formula (10) V ₂ decreases; the power loss shared by the thermal power generation is still loss ₁ ', but after connecting to the distributed power supply, the thermal power generation and the distributed power supply jointly provide electric energy to the load, and the power consumption of the load undertaken by the thermal power generation is less than before, so The unit power generation cost V ₁ of thermal power generation increases slightly. If the power loss of the active distribution network increases, Δloss is positive, and the increased power loss is borne by the distributed power generation, and the thermal power generation still bears the original power loss ₁ ', while the power consumption of the load is reduced.

By analogy, the correction of the power generation cost of various power sources caused by the introduction of demand-side response will not be repeated here.

Step 6. Solving the upper-level planning model: Use the above-mentioned revised power generation cost to correct the relationship curve between power generation utilization hours and unit capacity IRSP comprehensive cost, and solve the IRSP upper-level planning model to obtain the installed capacity of various power supply plans;

In step 6, the IRSP comprehensive cost includes the initial investment and construction cost, fuel cost, operation and maintenance cost and pollutant treatment cost of the power supply, wherein the initial investment and construction cost is a fixed part, and the fuel cost and operation and maintenance cost (the present invention unifies the two Defined as power generation cost) and pollutant treatment cost are variable parts. In addition, the above-mentioned costs should be calculated based on the cost price of the actual input rather than the market transaction price, which can more objectively reflect the real cost of various types of power investment and is conducive to the efficient use of resources from the perspective of the whole society. The expression of the unit capacity IRSP comprehensive cost Z _i of the above three types of power sources changing with the power generation utilization hour t is shown in formula (13):

Z _i =F _i +(V _i +ωβ _i )×t(13)

In the formula, i=1, 2, and 3 represent thermal power generation, photovoltaic power source, and interruptible load respectively; F _i is the annual value of the unit capacity construction cost of the i-type power source; V _i is the unit power generation cost of the i-type power source; β _i is the pollution control cost per unit electricity of the i-type power supply; ω is the weight, which depends on the degree of emphasis on environmental benefits. Among them, photovoltaic power generation does not need to invest in fuel costs, and its unit power generation cost V ₂ only includes equipment operation and maintenance costs. Interruptible load means that the user signs a contract with the power company, and the user cuts off part of the load during the peak load period, and the power company compensates the user for the removed load. Since power users lose the economic benefits that could have been generated by the load-shedding electricity, the contract signing fee for interruptible loads should be regarded as a fixed part of the IRSP comprehensive cost Z _i , and the compensation fee for load shedding should be regarded as a change in the IRSP comprehensive cost Z _i part.

In order to compare and select the power supply planning scheme with the lowest integrated IRSP cost Z _i per unit capacity, the present invention describes the relationship between the power generation utilization hours of three types of power sources and the integrated IRSP cost Z _i per unit capacity, as shown in Figure 1a. Among them, the abscissa is 8760 hours in a year, and the ordinate is the comprehensive cost Z _i of IRSP per unit capacity; Curves 1, 2 and 3 represent thermal power generation, photovoltaic power supply and interruptible load respectively, and the intercept is the comprehensive cost Z _i of IRSP per unit capacity The fixed part F _i ; the slope is the variable part V _i +ωβ _{i of the unit capacity IRSP comprehensive cost Z i .} Figure 1b is the predicted annual continuous load curve of a planning year in a region drawn using the data in Table 3. The abscissa is also 8760 hours throughout the year, and the ordinate is the active load power. The annual continuous load curve is based on the annual sequential load curve, regardless of its time sequence, which is obtained by rearranging the average load power L(t) per hour throughout the year in descending order. The annual hourly average load power L(t) satisfies formula (14):

L(t)＝L _y ×P _wk ×P _d ×P _h (t)(14)

In the formula, L _y is the annual peak load, P _wk is the percentage of the weekly peak load to the annual peak load, P _d is the percentage of the daily peak load to the weekly peak load, and P _h is the percentage of the hourly peak load to the daily peak load.

Table 3 Continuous Load Curve Parameters

When the weight of environmental benefits ω=1, the construction cost of thermal power generation is the highest, and the cost of the variable part is the lowest; the construction cost of photovoltaic power is slightly lower than that of thermal power generation, and the cost of the variable part is higher, so the slope is higher than that of thermal power generation; The fixed cost of interrupting the load is the lowest, but the variable cost is the highest among all types of power sources. If more attention is paid to environmental issues, the value of weight ω can be increased to obtain a more environmentally friendly planning scheme.

It can be seen from Fig. 1 that before time t ₁ , the integrated cost Z _i of the unit capacity IRSP of the interruptible load represented by curve 3 is lower than that of the other two types of power sources, so the area surrounded by P _C , P _MAX , and C represents the load The power consumption Q _L3 is solved through demand-side resources, that is, this part of the load is contracted to be transformed into an interruptible load, and the contracted capacity is P _MAX -P _C . Between t ₁ and t ₂ , the integrated cost Z _i of the unit capacity IRSP of the photovoltaic power source represented by curve 2 is lower than that of the other two types of power sources, so the area surrounded by P _B , P _C , C, and B represents the power consumption of the load The quantity Q _L2 is borne by the photovoltaic power source, and the installed capacity is P _C -P _B . After time _t2 , the unit capacity IRSP comprehensive cost Z _i of thermal power generation is the lowest, so the load power consumption Q _L1 represented by the area surrounded by PA, P _B , _B , and A is borne by thermal power generation, and the installed capacity is P _B - P _A . In order to simplify the problem research, the present invention assumes that the loads with power less than PA ₍ PA, _A , the continuous load curve and the area surrounded by the abscissa axis) are borne by the original power source of the active distribution network (before expansion planning).

Step 7. Solving the lower-level planning model: According to the installed capacity of the above-mentioned various power supply plans and the parameters set in step 3, use the improved PSO algorithm to solve the lower-level planning model to obtain the photovoltaic power supply with the minimum network loss and the interruptible load. Site selection and capacity planning;

Step 8. Judgment of iteration termination conditions: Judging whether the algorithm has reached the maximum number of iterations or found the optimal solution that meets the accuracy requirements. If so, jump out of the loop and output the optimal plan for IRSP regional power expansion planning; otherwise, go to step 5 and continue to iterate.

For this embodiment, the double-level planning results of the optimal allocation of regional power sources in the active distribution network provided by the present invention are as follows: the new installed capacity of thermal power generation is 2081.4kW, the new installed capacity of photovoltaic power supply is 1632.7kW, and the signed interruption capacity of interruptible loads is 757.9kW. See Table 4 for specific capacity data of selected nodes to configure photovoltaic power sources and interruptible loads. Compared with traditional planning, the comprehensive cost of IRSP per unit capacity has decreased by 503,900 yuan, a drop of 11.75%; the cost of power supply construction has been reduced by 738,000 yuan, a drop of 52.12%; The cost of property management decreased by 125,200 yuan, a drop of 18.42%, and the daily grid loss decreased by 1297.5kWh, a drop of 30.41%. The specific data of the comparison between the two-level planning method for the optimal configuration of the regional power supply of the active distribution network provided by the present invention and the traditional planning is shown in Table 5.

Table 4 Optimal configuration scheme of photovoltaic power supply and interruptible load

Table 5 The inventive method compares with traditional planning method result

Claims (7)

1. an active distribution network region electricity optimization configuration bi-level programming method, is characterized in that: it comprises the following step that order performs:

Step one, basic data obtain: the basic data comprising power supply type, grid structure, load level, electric parameter obtaining distribution system to be studied;

Step 2, model construction: the basic data utilizing step one to obtain builds the Bi-level Programming Models based on the region power extension problem of IRSP, and determines levels objective function and the constraint condition of this Bi-level Programming Models;

Step 3, algorithm parameter initialization: the parameter of initialization dual layer resist algorithm, the maximum iteration time of the PSO algorithm that setting improves is 50 times, particle populations number is n, particle code length is the node total number of accessible photo-voltaic power supply and interruptible load in distribution system, Studying factors C1=2, C2=1.732, and determine stopping criterion for iteration;

Step 4, initial network loss solve: assuming that higher level's substation capacity is sufficient, do not consider photo-voltaic power supply and interruptible load, Energy loss when utilizing lower floor's plan model calculating active distribution network load all to be powered by higher level transformer station, this Energy loss is born by higher level transformer station and thermal power generation;

Step 5, the correction of unit quantity of electricity cost of electricity-generating: utilize lower floor's plan model to solve the Energy loss obtained to share to all kinds of power supply, to revise the cost of electricity-generating of all kinds of power supply by above-mentioned;

Step 6, upper strata plan model solve: utilize above-mentioned revised cost of electricity-generating correction gas-to electricity hour and unit capacity IRSP integrated cost relation curve, and solve IRSP upper strata plan model, to obtain the installed capacity of all kinds of power source planning;

Step 7, lower floor's plan model solve: according to the parameter set in the installed capacity of above-mentioned all kinds of power source planning and step 3, utilize the PSO algorithm improved to solve lower floor's plan model, obtain and meet the photo-voltaic power supply of loss minimization and the addressing constant volume scheme of interruptible load;

Step 8, stopping criterion for iteration differentiate: whether evaluation algorithm reaches maximum iteration time or search the optimum solution meeting accuracy requirement, if then jump out circulation, export IRSP region Expansion Planning of power plants optimal case; Otherwise go to step five, proceed iteration.

2. active distribution network region according to claim 1 electricity optimization configuration bi-level programming method, it is characterized in that: in step one, described power supply type is selected newly-built thermal power generation, investment photo-voltaic power supply and is signed interruptible load contract, represents the power supply type that the source substation of power distribution network, distributed power source and Demand-side resource these three kinds is different respectively.

3. active distribution network region according to claim 1 electricity optimization configuration bi-level programming method, it is characterized in that: in step 2, the mathematic(al) representation of the objective function of described upper strata plan model is:

$\min f=\underset{i=1}{\overset{I}{\&Sigma;}}(F_i{C}_i+V_i{C}_i{H}_i)+\underset{i=1}{\overset{I}{\&Sigma;}}{H}_i{C}_i{\mathrm{\&beta;}}_i---\left(1\right)$

In formula, i is power type, i=1, and 2,3 represent thermal power generation, photo-voltaic power supply, interruptible load respectively; F _ifor such power supply unit capacity construction cost year value, unit/kW; C _ifor such power supply installation total volume, kW; V _ifor such power supply unit quantity of electricity cost of electricity-generating, unit/kWh; H _ifor such power supply annual utilization hours, hour; β _ifor the pollutant control cost of such power supply unit quantity of electricity, unit/kWh; F _ican be expressed as further:

$F_i=a_i\&CenterDot;\frac{r_0\&CenterDot;{(1+r_0)}^m}{{(1+r_0)}^m-1}---\left(2\right)$

In formula, a _ifor the unit capacity cost of such power supply, unit/kW; r ₀for rate of discount; M is the operation time limit of such power supply, year;

The constraint condition of upper strata plan model is:

$\underset{i=1}{\overset{I}{\&Sigma;}}{C}_i{H}_i\&GreaterEqual;{\&Integral;}_0^{8760}L\left(t\right)\&CenterDot;dt---\left(3\right)$

In formula, L (t) is annual average load power hourly.

4. active distribution network region according to claim 1 electricity optimization configuration bi-level programming method, it is characterized in that: in step 2, the mathematic(al) representation of the objective function of described lower floor's plan model is:

minf＝loss(4)

Constraint condition is:

$P_{pv}+P_{IL}-P_i=U_i\underset{j=1}{\overset{n}{\&Sigma;}}{U}_j(G_{ij}{\mathrm{cos\&delta;}}_{ij}+B_{ij}{\mathrm{sin\&delta;}}_{ij})---\left(5\right)$

$Q_{pv}+Q_{IL}-Q_i=U_i\underset{j=1}{\overset{n}{\&Sigma;}}{U}_j(G_{ij}{\mathrm{sin\&delta;}}_{ij}-B_{ij}{\mathrm{cos\&delta;}}_{ij})---\left(6\right)$

U _imin≤U _i≤U _imax(7)

0≤I _j≤I _jmax(8)

S _j≤S _jmax(9)

In formula, loss is Energy loss; N is system node number; P _pv, Q _pv, P _iL, Q _iLbe respectively the meritorious and reactive power of distributed photovoltaic and interruptible load injection node i; P _i, Q _ibe respectively the meritorious of node i and load or burden without work; G _ij, B _ijbe respectively the corresponding element in bus admittance matrix; U _ifor the voltage magnitude of node i, U _imin, U _imaxbe respectively permission upper voltage limit and the lower limit of node i; I _jfor the current amplitude of branch road j, I _jmaxfor the electrical current heat of branch road j stablizes the upper limit; S _jfor the applied power of branch road j, S _jmaxfor the applied power upper limit of branch road j.

5. active distribution network region according to claim 1 electricity optimization configuration bi-level programming method, is characterized in that: in step 5, described every class power supply unit quantity of electricity cost of electricity-generating V _ifor:

$V_i=\frac{S_{pdi}\&times;({\mathrm{load}}_{dem}+{\mathrm{loss}}_i)}{{\mathrm{load}}_{dem}}---\left(10\right)$

In formula, S _pdiit is the unit quantity of electricity cost of electricity-generating of this power supply originally; Load _demit is the load power consumption that such power supply is born; Loss _iit is the Energy loss that such power supply is born.

6. active distribution network region according to claim 1 electricity optimization configuration bi-level programming method, it is characterized in that: in step 5, described to be shared by Energy loss to the method for all kinds of power supply be carry out in two steps, and the first step will not shared to original electricity provider containing active distribution network Energy loss during distributed power source; The Energy loss variable quantity that second step causes after being accessed by distributed power source is shared to distributed power source;

Δloss＝loss ₁-loss′ ₁(11)

loss ₂＝Δloss(12)

In formula, Δ loss is loss ₁with loss ₁' difference, loss ₁for the total Energy loss after access distributed power source, loss ₁' be not containing Energy loss during distributed power source; Loss ₂for the Energy loss that distributed power source is shared.

7. active distribution network region according to claim 1 electricity optimization configuration bi-level programming method, it is characterized in that: in step 6, described IRSP integrated cost comprises the initial outlay construction cost of power supply, fuel cost, operation expense and pollutant control cost, wherein initial outlay construction cost is fixed part, and fuel cost, operation expense and pollutant control cost are variation part; Above-mentioned three class power supplys are with the unit capacity IRSP integrated cost Z of gas-to electricity hour t change _iexpression formula such as formula shown in (13):

Z _i＝F _i+(V _i+ωβ _i)×t(13)

In formula, i=1,2,3 represents thermal power generation, photo-voltaic power supply, interruptible load respectively; F _iit is the i-th class power supply unit capacity construction cost year value; V _iit is the i-th class power supply unit quantity of electricity cost of electricity-generating; β _iit is the i-th class power supply unit quantity of electricity pollutant control cost; ω is weight, depending on environmental benefit by attention degree.