Power system multi-target collaborative planning method considering interconnection and mutual aid of multi-regional power grids

Abstract

The invention discloses a power system multi-target collaborative planning method considering interconnection and mutual aid of multi-regional power grids. The method comprises the following steps: constructing a power system reliability index; constructing a power supply margin index of the power system; constructing a double-layer optimization model; and solving the double-layer optimization model. According to the method, establishment of multi-target collaborative planning of the multi-regional power grid interconnection and mutual assistance power system is considered, an upper-layer model is a multi-target model for maximizing power system reliability and power supply margin indexes, a lower-layer model is a target for minimizing the comprehensive cost of power system capacity planning and operation scheduling, and a double-layer optimization model is constructed; the harmony search algorithm is used for solving the double-layer model, the performance is excellent in the aspect of searching the globally optimal solution, the speed of the harmony search algorithm is far higher than that of a traditional method, information can be effectively explored and utilized in a search space, and the globally optimal solution can be found more quickly; the reliability and the multi-region interconnection characteristic of the novel power system are considered, cross-region power grid interconnection is promoted from the perspective of power system source grid load storage integrated planning, and construction of the novel power system is facilitated.

Description

Technical Field

[0001] The present invention relates to a collaborative planning method, and in particular to a multi-objective collaborative planning method for an electric power system taking into account the interconnection of multi-regional power grids. Background Art

[0002] At present, power system planning mainly focuses on single power source planning or power grid planning. Some results have studied the capacity planning methods of wind, light, water, thermal and other power sources in power systems, as well as multi-regional power grid planning methods, but have failed to consider the coordinated planning of power systems with integrated source, grid, load and storage under multi-regional interconnection.

[0003] Regarding the research on power system planning models, most of the existing studies have established a single-layer optimization model with minimizing investment costs as the objective function. Some studies have considered the power system reliability constraint in the constraints in the planning model. However, the traditional planning model ignores the fact that in the planning of new power systems, the flexibility requirements are higher and the time scale requirements are smaller, and a more refined model needs to be established. Therefore, the patent of the present invention combines the above two points and considers the establishment of a multi-objective collaborative planning of a power system with multi-regional power grid interconnection. The upper model is a multi-objective model that maximizes the reliability and power supply margin indicators of the power system, and the lower model is a double-layer optimization model with the goal of minimizing the comprehensive cost of power system capacity planning and operation scheduling.

[0004] When faced with the problem of solving a two-layer model, traditional research methods often rely on KKT conditions and Lagrangian functions to transform the objective function, simplifying the two-layer optimization problem into a single-layer problem for easy processing. However, in practice, most models are non-convex functions and cannot be transformed through KKT. Secondly, the solution process is often very complicated. Summary of the invention

[0005] In order to solve the defects existing in the prior art, the present invention discloses a multi-objective collaborative planning method for a power system considering the interconnection and mutual assistance of multi-regional power grids, and its technical solution is:

[0006] The multi-objective collaborative planning method of power system considering the interconnection of multi-regional power grids is characterized by:

[0007] Step a: constructing power system reliability indicators;

[0008] Step b: constructing the power supply margin index of the power system;

[0009] Step c, constructing a two-layer optimization model;

[0010] Step d: solving the two-layer optimization model.

[0011] The present invention also discloses a non-volatile storage medium, characterized in that the non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the above method.

[0012] The present invention also discloses an electronic device, characterized in that it comprises a processor and a memory; the memory stores computer-readable instructions, and the processor is used to execute the computer-readable instructions, wherein the computer-readable instructions execute the above method when they are executed.

[0013] Beneficial Effects

[0014] The present invention considers establishing a multi-objective collaborative planning of an electric power system with interconnected power grids in multiple regions. The upper model is a multi-objective model that maximizes the reliability and power supply margin indicators of the electric power system, and the lower model is aimed at minimizing the comprehensive cost of power system capacity planning and operation scheduling. A double-layer optimization model is constructed, and the harmony search algorithm is used to solve the double-layer model. It not only performs well in finding the global optimal solution, but its speed far exceeds that of traditional methods. It can effectively explore and utilize information in the search space, thereby finding the global optimal solution more quickly. Considering the reliability of the new electric power system and the characteristics of multi-regional interconnection, it starts from the perspective of integrated planning of power system source, grid, load and storage, which is helpful to promote cross-regional power grid interconnection and is conducive to the construction of a new electric power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The block diagram of the two-layer optimization model construction;

[0016] Figure 2 Flowchart of calculation of hybrid solution algorithm. DETAILED DESCRIPTION

[0017] The present invention proposes a multi-objective collaborative planning method for a power system that considers the interconnection of multiple regional power grids. The method mainly includes the steps of a. constructing a reliability index of the power system, b. constructing a power supply margin index of the power system, c. constructing a two-layer optimization model, and d. solving the two-layer optimization model. The specific contents are as follows:

[0018] a. Construction of reliability index of power system For the calculation of reliability index, the present invention uses three typical reliability indexes to derive, namely: annual failure rate λ, instantaneous failure rate μ, and annual power outage time ω.

[0019] a1. Annual failure rate λ. Annual failure rate refers to the frequency of equipment or system failures within a year.

[0020]

[0021] Where: M represents the set of multi-regional power systems, m represents the m-th multi-regional power system; K represents the set of subject types in the multi-regional power system, k represents the k-th subject type; U m,k represents the number of subject types k in the self-balancing unit m of the power system. The fault events and occurrence probabilities in the system are statistically based on the subject units. For subjects of the same type k, it is assumed that the probability of fault occurrence is the same; λ m,k represents the annual failure rate of main body type k in the power system self-balancing unit m.

[0022] a2. Instantaneous failure rate: The instantaneous failure rate refers to the conditional probability of equipment failure at a specific moment or within a short period of time.

[0023]

[0024] Where: μ m,k represents the instantaneous failure rate of the main body type k in the power system self-balancing unit m.

[0025] a3. Annual power outage time. Annual power outage time refers to the total power outage time caused by power system or equipment failure or other reasons within one year.

[0026]

[0027] Where: m,k It represents the annual outage time of main type k in the power system self-balancing unit m.

[0028] b. Construction of power system power supply margin indicators.

[0029] On the basis of considering both active and reactive power, a power supply margin index based on probability index is proposed. Since there are probabilistically changing power generation elements and loads in the system, a multi-state power generation-load model is used here to calculate the system margin index IS. The power supply margin index of the self-balancing unit m is defined as a cumulative index, considering the probability of each power generation-load state, as well as the active and reactive network losses inside the self-balancing unit m in each state. The power supply margin index of the whole system is the sum of the active and reactive power supply margin indicators of all multi-regional power systems.

[0030] It should be noted that the total output power of DG is usually lower than the total power consumption of users in the distribution network. This means that in a power system containing multiple multi-regional power systems, when the power generation is greater than the power consumption in the multi-regional power system, other multi-regional power systems will have energy storage units to store excess power, and vice versa. In order to optimize the power balance of the system, the power supply margin indicator here takes into account the two situations of insufficient and surplus power generation in the multi-regional power system.

[0031]

[0032] Where: Y represents the power generation-load state set; and They represent the active and reactive power of the generation / storage unit in the self-balancing unit m in state y respectively; and They represent the active and reactive power requirements of the self-balancing unit m in state y respectively; τ y represents the probability of the self-balancing unit m being in the power generation-load state y; the present invention assumes that in the current state, the active and reactive network losses of the multi-regional power system are 5% of the active and reactive loads respectively; represents the maximum apparent power in the self-balancing unit m; z M Represents a Boolean variable. When power generation is insufficient, z M Assign a value of -1; when there is surplus power generation, z M Assign a value of 1. From the mathematical properties, I t The value of increases with the increase of margin: when there is no generator set in the system, I t = -1; I t As it approaches 0, it indicates that the system supply and demand match is better; I t The larger the value, the higher the system power supply margin. In terms of system reliability, I t The larger the value, the higher the margin and the stronger the system reliability. However, there is a contradiction between the reliability target and the economic target of the lower model, that is, the reliability target must meet certain economic constraints. Therefore, the present invention is for I t The value of I sets the maximum value that can be economically tolerated. max (I max >1), when the calculated value is greater than or equal to I max When t Directly assign the value to Imax. In order to coordinate the analysis of the reliability index and the system power supply margin index, the present invention assigns the value to Imax. t Processing is performed and the system margin index value I of the tth year is obtained Mt . with I t On the contrary, I Mt The smaller the value, the higher the system power supply margin.

[0033] c. Constructing a two-layer optimization model. The present invention designs a two-layer optimization model for collaborative planning of power systems that takes into account multi-regional interconnection and mutual assistance, and iteratively optimizes the upper model and the lower model through an iterative algorithm to establish a final planning scheme. Figure 1 For the framework diagram of the two-layer optimization model for coordinated planning of power systems considering multi-regional interconnection, see Figure 1 shown.

[0034] c1. Upper model.

[0035] The upper model of the multi-regional planning model for multi-regional interconnected power systems focuses on how to achieve the optimal goal of system reliability under the constraints of system flow, capacity configuration, power balance, etc. when multiple power systems are interconnected across regions. According to the model design, each multi-regional power system is regarded as a node in the upper model. When the system is in operation, it is necessary to reduce the load, export power, import power, etc. on the node, which is based on the multi-regional power system. The optimization basis of the upper model, that is, the configuration schemes of DG, capacitors, grid structure, etc. in the multi-regional power system and the internal flow balance during operation, are implemented in the lower model. In order to better reflect the goals of the upper model, we introduced the RMI index, which covers two sub-indicators of reliability and power supply margin, which is the optimization goal pursued by the upper model. Through the setting of RMI, the two model target indicators are normalized into a single-objective optimization problem. The derivation of RMI is as follows:

[0036]

[0037] Where: I Rt and I Mt They are system reliability index and system power supply margin index respectively, and T is the total number of years in the planning period.

[0038] Since the dimensions of the above three indicators are different, in order to facilitate subsequent processing and analysis, the indicators need to be normalized first. The normalization method is as follows:

[0039]

[0040] Where: I Rt represents the value of the normalized reliability index in the tth year; max , μ max and ω max They represent the annual failure rate of the system within the operating range λ t , system instantaneous failure rate μ t 、System annual power outage timeω t Maximum value; λ min , μ min and ω min They represent the annual failure rate of the system based on data statistics λ t , system instantaneous failure rate μ t 、System annual power outage timeω t The minimum probability value; w1, w2 and w3 are the weights of the three indicators, which can be adjusted according to the importance of the system to the three indicators. Rt The smaller it is, the higher the system reliability is.

[0041] From the above analysis, it can be seen that the annual failure rate of the whole system λ t, system instantaneous failure rate μ t 、System annual power outage timeω t , the margin index IS is closely related to the configuration of components such as DG, grid structure, capacitors, etc. in each multi-regional power system, as well as the connection and flow distribution between multi-regional power systems. For example, the more sufficient the power generation in the multi-regional power system, the shorter the power outage time of users will be. At the same time, due to the reduction of energy exchange between multi-regional power systems, the line failure rate will also be reduced; for example, the unreasonable configuration of the grid structure in the self-balancing unit will lead to low utilization of DG, prolonged power outage time for users, and excessive load on some lines, which increases the risk of failure. It can be seen from this that the objective function of the upper model needs to be realized under the constraints of the planning and operation stages between multi-regional power systems, and also needs the support of the optimization scheme of the lower model.

[0042] Constraints are mainly used to ensure the reasonable implementation of cross-regional optimization planning and operation of power systems, including power balance and voltage stability at the multi-regional power system end, as well as capacity constraints and flow constraints of regional grid architecture. In the upper model system, each multi-regional power system can be regarded as a system node to construct constraints.

[0043] 1) Power balance constraints in multi-regional power systems

[0044] In addition to power provided by internal DG and energy storage, multi-regional power systems can also choose external power input and power output according to their own load conditions, thereby ensuring the power balance of the multi-regional power system to maintain system safety and stability.

[0045]

[0046] Where: G represents the generation type set of DG, G = {g / g = g x ,x=1,2,3,…}; g represents the g-th generation type of DG; T represents the total number of years in the planning period; represents the initial annual capacity of generation type g in the DG units; represents the newly added capacity of power generation type g in year t; H t H represents the available energy storage capacity in year t. t Under certain conditions, charge and discharge conversion can be achieved. t When it is a positive number, it indicates the discharge capacity of the energy storage. t When it is a negative number, it indicates the storage capacity of energy storage; P t in and P t out Respectively represent the input capacity of other multi-regional power systems and the output capacity of multi-regional power systems; D t Indicates the load actually absorbed in the multi-regional power system; It represents the load reduction due to demand response and load reduction in the multi-regional power system. t in and P t out It is also subject to the line capacity constraints between the multi-regional power system and other multi-regional power systems, which will be explained in detail when explaining the line capacity constraints.

[0047] 2) System node voltage constraints

[0048]

[0049] Where: V min ,V max Respectively represent the maximum and minimum voltage values ​​allowed for each multi-region power system (node); V t,b′ represents the voltage value of any load segment b' of the multi-regional power system in year t; assuming that B' is a time period set, B'={b' / b'=b' x ,x=1,2,3,…},b' x Indicates the xth time period.

[0050] 3) Power flow constraints

[0051] The line capacity between the self-balancing units m and m' is limited, and the active and reactive power that can be carried should be constrained.

[0052]

[0053] Where: P m , Q m and |V m | respectively represent the active power, reactive power injection and voltage amplitude of the port of the self-balancing unit m under normal conditions; G mm' and A mm' Respectively represent the real and imaginary parts of the system admittance matrix under normal conditions; θ mm' Represents the voltage phase angle difference between the self-balancing units m and m'.

[0054] 4) Line capacity constraints

[0055]

[0056] Where: represents the total capacity of the line between self-balancing units m and m' in year t; represents the total capacity of the line between self-balancing units m and m' in the initial year; represents the increase in line capacity between self-balancing units m and m' in year t; Indicates the maximum carrying capacity of the line between self-balancing units m and m'.

[0057] 5) Reliability constraints

[0058] Considering the actual operating status and economy of the system, the reliability indicators involved in the model objective function should be limited to a certain range of values.

[0059]

[0060] c2. Lower model.

[0061] The goal of the lower-level planning model is to achieve the optimal configuration of DG, grid architecture, energy storage, switches and other components in the multi-regional power system while minimizing the total investment during the planning period. Therefore, the model takes the minimum total cost during the planning period as the goal, and comprehensively considers the investment and operating costs of distributed generation, the investment costs and construction costs required for line upgrades, the investment costs of energy storage units, the load reduction costs and the demand response costs, as shown below.

[0062]

[0063] The total investment cost of a component is expanded using an auxiliary function:

[0064]

[0065] Where: I represents the node set, I = {i / i = i x ,x=1,2,3,…},i x represents the xth node; G represents the generation type set of DG, G={g / g=g x ,x=1,2,3,…}, same as the above model; g represents the g-th generation type of DG; T is the total number of years in the planning period, same as the above model; t represents the t-th year in the planning period; represents the unit capacity investment cost of DG units of power generation type g; represents the newly added DG unit capacity of generation type g in node i in year t; represents the unit capacity operating cost of the DG unit of generation type g; represents the DG operating capacity of generation type g in node i in year t; h g,i,t represents the annual operation time of the power generation type g in node i in year t; r represents the discount rate; w represents the optional line type when the transmission line is expanded; Represents the unit length line investment and construction cost of line type w; π ij,t represents the number of newly added line loops between nodes i and j in year t; l ij Represents the length between node i and node j; represents a Boolean variable, It indicates that there is expansion between node i and node j in year t, and vice versa; represents the unit capacity operating cost of the line type w; represents the actual operating capacity between nodes i and j in year t; CF S , CO S They represent the unit capacity investment cost and unit operating cost of the energy storage unit respectively; represents the newly added energy storage capacity in node i in year t; represents the total amount of electricity flowing between the system and the energy storage unit in node i in year t; CF Cap Represents the unit capacity investment cost of the parallel capacitor; Q Cap Indicates the rated capacity of each parallel capacitor. The capacitors in this article are all set to the same model. represents the number of parallel capacitors installed at node i in year t; C U represents the system outage cost, which is composed of the demand response fee C DR and load reduction cost C cut Composition; B is the peak load segment set, B = {b / b = b x ,x=1,2,3,…},b x represents the xth peak load segment; c DR represents the unit cost of implementing demand response in the system; represents the load reduced by demand response in peak load period b in year t; z b,t is a Boolean variable, z b,t =1, it means that the peak load period b in year t has achieved load reduction, and vice versa; c cut It represents the unit compensation fee paid to users for directly reducing user load; represents the amount of electricity directly reduced by users during peak load period b in year t; z′ b,t is a Boolean variable, z′ b,t =1, it means that the peak load segment b in year t has achieved direct load reduction, and vice versa; N represents the collection of all devices in the system, N = {n / n = n x ,x=1,2,3,…},n x Indicates the nth device; T n Indicates the life span of equipment n; t n Indicates the year when the equipment is put into use, that is, the number of years counted from the initial year of the planning period.

[0066] The lower model mainly considers the planning and operation of multi-regional power systems, involving system components such as DG, grid, and energy storage. Therefore, when setting constraints, it mainly starts from meeting the basic requirements of planning and operation, as well as the operating status that each component and multi-regional power system needs to achieve. Based on the above objective function, the main constraints to be considered are DG capacity constraints, line capacity constraints, energy storage constraints, flow constraints, voltage constraints, and power quality constraints.

[0067] 7) DG capacity constraints

[0068] The active power and reactive output of the DG unit are limited by its maximum capacity and minimum capacity. At the same time, the installable DG capacity of node i is also constrained, as shown in the following formula:

[0069]

[0070] Where: They represent the upper and lower limits of active output of generation type g in the DG unit respectively; They represent the upper and lower limits of reactive power output of generation type g in the DG unit respectively; represents the capacity of generation type g in node i in the initial year; Indicates the total capacity of DG that can be installed on node i.

[0071] 8) Line capacity constraints

[0072] The power flow through any distribution line must comply with the capacity limit of the line, and the upgrade and expansion of the line needs to take into account the constraints of the wiring space.

[0073]

[0074] Where: represents the newly added capacity of feeder ij in year t; represents the reactive power flow on feeder ij in year t; represents the power angle of the current from node i to node j in year t. M is a sufficiently large number so that when hour, is 0; when When , the value of M is large enough to ensure that there is enough space to select new lines. Boolean variable As a "switch" of continuous constraints, produce a capacity greater than or equal to the maximum capacity allowed by the line upgrade.

[0075] 9) Energy storage constraints

[0076] Since the energy storage unit in the multi-regional power system must be able to charge and discharge to track the rapid changes in load, and have a large energy storage capacity, but is limited by its own technical conditions, the charging and discharging capacity and rate of the energy storage in actual operation must meet the following constraints:

[0077]

[0078] Where: and are the charging and discharging amounts of the energy storage unit in time period b' respectively; to facilitate marking the state of charge (SOC) of the energy storage unit, the t-th year is divided into several load segments, assuming that B' is a time period set, B'={b' / b'=b' x ,x=1,2,3,…},b' x represents the xth time period; η c and η d are the charging and discharging efficiency of the energy storage unit in load segment b'; H i,b′ Indicates the SOC of the energy storage unit during load segment b'; H max and H min are the upper and lower limits of the SOC allowable range respectively; ν b′ represents the charging and discharging rate of the energy storage unit in the load segment b'; ν max Indicates the maximum limit allowed for the charging and discharging rate of the energy storage unit.

[0079] 10) Power flow constraints

[0080]

[0081] Where: P i , Q i and |V i | respectively represent the active power, reactive power injection and voltage amplitude of node i under normal conditions; G ij and A ij Respectively represent the real and imaginary parts of the system admittance matrix under normal conditions; θ ij It represents the voltage phase angle difference between node i and node j.

[0082] 11) Reserve capacity constraints

[0083]

[0084] Where: g represents the forced outage rate of DG generation mode g; δ represents the reserve capacity rate; represents the maximum load of node i in year t.

[0085] 12) Power quality constraints

[0086] The access of DG will significantly affect the node voltage of the distribution network. If the configuration of DG is unreasonable, the voltage stability will be greatly impacted, which will greatly limit the ability of the distribution network to withstand load growth. Therefore, voltage stability is very important for the safe and stable operation of the system. The power quality constraint mainly considers the voltage stability VSI (voltage stability index), which is expressed by the index ξ. VSI index ξ of branch I l It can be expressed as:

[0087]

[0088] ξ l ≤V l∈L ij (37)

[0089] Where: R ij and X ij are the resistance and reactance of branch I respectively; P j and Q j is the active power and reactive power of the receiving end point j of branch I; L ij is the total number of system branches. l The branch corresponding to the maximum value in is called the weakest branch of the system. When the system is stable, the value corresponding to the weakest branch must be less than 1, and the voltage collapse of the system always starts from the weakest branch in the system. l The smaller the voltage, the better the stability; the larger the voltage, the worse the stability. l When it approaches 1, the system voltage collapses. This constraint ensures that all ξ values ​​are less than a fixed value V<1, thereby ensuring the stability of the system voltage.

[0090] d. Solution of the two-layer optimization model.

[0091] The model of the present invention is essentially a type of nonlinear hybrid two-level planning problem: the upper level is a nonlinear hybrid optimization problem for cross-regional power supply and demand and grid topology planning; the lower level is an integer optimization problem for DG, energy storage, capacitor site selection and capacity determination, and grid topology planning.

[0092] To solve the above complex model, a harmony search hybrid algorithm with embedded optimal probability power flow point estimation is designed. Its calculation process is as follows: Figure 2 shown.

[0093] The basic idea of ​​the algorithm is to use the heuristic strategy and strong global search ability of the harmony search algorithm (HAS) as a tool for solving the main problem of the upper layer of the model. For the large number of feasible planning schemes generated, the key lies in the uncertainty of the DREG output power and the uncertainty of the probability flow distribution caused by the location and capacity of components such as DG, energy storage, and capacitors. If the DREG output and component configuration problems are simulated based on the traditional Monte Carlo simulation, there are often problems of large calculation amount and low efficiency. Therefore, the present invention adopts a three-point estimation method to calculate the optimal probability flow of the system under each scenario and component configuration scheme. This method gets rid of the linearization and cumulative calculation of the flow equation, but for each probability random variable, it constructs 3 estimation points using its high-order moment, and then solves the objective function value at each estimation point through a deterministic optimal flow model and algorithm, and then obtains the statistical characteristic quantity of the objective function. For example, if the problem contains n random variables, 3n estimates are required to obtain the probability density distribution of the output variable. Since this method realizes the transformation of the probability optimization problem containing large-scale uncertain information into the inference problem of the original feature quantity based on small-scale deterministic information, it will significantly reduce the amount of calculation for solving the lower-level model compared with Monte Carlo simulation. The solution steps of the M IPEI model are as follows:

[0094] 1) Initialization of algorithm parameters. Including harmony memory size HMS, harmony memory selection probability HMCR, pitch fine-tuning probability PAR, pitch fine-tuning bandwidth bw and maximum creation times T max .

[0095] 2) According to the number of candidate nodes of DG, energy storage and capacitor in the system, determine the number of tones required in the unit harmony sequence. Each candidate planning scheme corresponds to a harmony in the HSA memory bank, which is encoded in integer form, and its value corresponds to the configuration capacity of DG, energy storage and capacitor in the system. Randomly generate feasible solution vectors of HMS group.

[0096] 3) Based on the three-point estimation method, the expected output power of DREG and the probability power flow distribution of the self-balancing unit m under normal operation of each planning scheme are calculated;

[0097] 4) Combined with the probabilistic power flow distribution during the operation phase, calculate and record the corresponding investment costs of each multi-regional power system and the configuration scheme of system components such as DG, energy storage, and capacitors;

[0098] 5) The expected output power and configuration scheme of DREG of the lower model are fed back to the upper model as system operation parameters, and the system reliability index and the upper objective function value F(·) are calculated accordingly.

[0099] 6) Use the linear sorting method to calculate the fitness value J of each individual in the current harmony memory bank and sort them in descending order.

[0100]

[0101] Where: ζ is the adjustment coefficient; x h Y is the position of the individual after being sorted by the upper layer objective function value; HMS is the number of model constraints.

[0102] 7) Compare the above harmony individuals with the individuals with the lowest fitness value in the original memory bank. If J(x h,new )>J(x h,hold ), the inferior harmony in the original library is replaced by a better harmony, and the harmony memory library is updated.

[0103] 8) Determine whether the maximum number of creation times T has been reached max If yes, output the current optimal harmony as the final planning scheme; otherwise, perform harmony memory library learning, pitch fine-tuning, and pitch random selection according to the HSA rules to generate a new harmony library, and return to step 3).

[0104] The technical solution of the present invention takes the regional power system as the main body, considers the interconnection and coupling of power systems in different regions, and proposes a multi-objective collaborative planning model for a power system with multi-regional power grid interconnection. The upper model is a multi-objective model that maximizes the reliability and power supply margin indicators of the power system, and the lower model takes the minimum comprehensive cost of power system capacity planning and operation scheduling as the goal, constructs a two-layer optimization model, and uses the harmony search algorithm to solve the two-layer model, so that the RMI index of the power system under multi-regional interconnection is maximized while minimizing the planning cost, better recommends the construction of multi-regional power grid interconnection, and promotes the transformation of my country's power system.

[0105] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A multi-objective collaborative planning method for power systems considering the interconnection of multiple regional power grids is characterized by: Step a: constructing power system reliability indicators; Step b: constructing the power supply margin index of the power system; Step c, constructing a two-layer optimization model; Step d: solving the two-layer optimization model.

2. The multi-objective collaborative planning method for a power system considering the interconnection of multiple regional power grids according to claim 1 is characterized by: The construction of the reliability index of the power system includes the following contents: For the calculation of the reliability index, three typical reliability indexes are used for derivation, namely: annual failure rate λ, instantaneous failure rate μ, annual power outage time ω, a1. Annual failure rate λ: Annual failure rate refers to the frequency of equipment or system failures within one year; Where: M represents the set of multi-regional power systems, m represents the m-th multi-regional power system; K represents the set of subject types in the multi-regional power system, k represents the k-th subject type; U m,k represents the number of subject types k in the self-balancing unit m of the power system. The fault events and occurrence probabilities in the system are statistically based on the subject units. For subjects of the same type k, it is assumed that the probability of fault occurrence is the same; λ m,k represents the annual failure rate of main body type k in the self-balancing unit m of the power system; a2. Instantaneous failure rate: Instantaneous failure rate refers to the conditional probability of equipment failure at a specific moment or within a short period of time; Where: μ m,k represents the instantaneous failure rate of the subject type k in the self-balancing unit m of the power system; a3. Annual power outage time: Annual power outage time refers to the total power outage time caused by power system or equipment failure or other reasons within one year: Where: m,k It represents the annual outage time of main type k in the power system self-balancing unit m.

3. The multi-objective collaborative planning method for a power system considering the interconnection of multiple regional power grids according to claim 1 is characterized by: The power system power supply margin index construction includes the following contents: A multi-state generation-load model is used to calculate the system margin index IS; the power supply margin index of the self-balancing unit m is defined as a cumulative index, taking into account the probability of each generation-load state, as well as the active and reactive network losses within the self-balancing unit m in each state; the power supply margin index of the entire system is the sum of the active and reactive power supply margin indicators of all multi-regional power systems; the power supply margin index takes into account the insufficient and surplus power generation in the multi-regional power system: Where: Y represents the power generation-load state set; and They represent the active and reactive power of the generation / storage unit in the self-balancing unit m in state y respectively; and They represent the active and reactive power requirements of the self-balancing unit m in state y respectively; τ y represents the probability of the self-balancing unit m being in the generation-load state y; represents the maximum apparent power in the self-balancing unit m; z M Represents a Boolean variable. When power generation is insufficient, z M Assign a value of -1; when there is surplus power generation, z M Assign 1; from the mathematical properties, I t The value of increases with the increase of margin: when there is no generator set in the system, I t = -1; I t As it approaches 0, it indicates that the system supply and demand match is better; I t The larger the value, the higher the system power supply margin. t The value of I sets the maximum value that can be economically tolerated. max , when the calculated value is greater than or equal to I max When t Directly assign the value to Imax; in order to coordinate the reliability index with the system power supply margin index, by t Processing is performed and the system margin index value I of the tth year is obtained Mt ;with I t On the contrary, I Mt The smaller the value, the higher the system power supply margin.

4. The multi-objective collaborative planning method for a power system considering the interconnection of multiple regional power grids according to claim 1 is characterized by: The upper model in the two-layer optimization model includes the following contents: through the setting of RMI, the two model target indicators are normalized into a single-objective optimization problem. The derivation of RMI is as follows: Where: I Rt and I Mt are the system reliability index and system power supply margin index respectively, and T is the total number of years in the planning period; Perform normalization of indicators. The normalization method is as follows: Where: I Rt represents the value of the normalized reliability index in the tth year; max , μ max and ω max They represent the annual failure rate of the system within the operating range λ t , system instantaneous failure rate μ t 、System annual power outage timeω t Maximum value; λ min , μ min and ω min They represent the annual failure rate of the system based on data statistics λ t , system instantaneous failure rate μ t 、System annual power outage timeω t The minimum probability value; w1, w2 and w3 are the weights of the three indicators, which are adjusted according to the importance of the system to the three indicators. Rt The smaller it is, the higher the system reliability is; From the above analysis, it can be seen that the annual failure rate of the whole system λ t , system instantaneous failure rate μ t 、System annual power outage timeω t ,The margin index IS is closely related to the DG, grid structure, capacitor element configuration within each multi-regional power system, and the connection and power flow distribution between multi-regional power systems.

5. The method for multi-objective collaborative planning of a power system considering the interconnection of multiple regional power grids according to claim 4 is characterized by: The objective function of the upper model needs to be implemented under the constraints of the planning and operation stages between the multi-regional power systems, and also needs the support of the optimization scheme of the lower model; the constraints are mainly used to ensure the reasonable implementation of cross-regional optimization planning and operation of the power system, including power balance and voltage stability at the multi-regional power system end, and capacity constraints and flow constraints of the regional grid architecture; in the upper model, each multi-regional power system can be regarded as a system node to construct constraints: 1) Power balance constraints in multi-regional power systems In addition to the power provided by internal DG and energy storage, the multi-regional power system selects external power input and power output according to its own load conditions, thereby ensuring the power balance of the multi-regional power system to maintain system safety and stability; Where: G represents the generation type set of DG, G = {g / g = g x ,x=1,2,3,...}; g represents the g-th generation type of DG; T represents the total number of years in the planning period; represents the initial annual capacity of generation type g in the DG units; represents the newly added capacity of power generation type g in year t; H t represents the available energy storage capacity in year t; H t Under certain conditions, charge and discharge conversion can be achieved. t When it is a positive number, it indicates the discharge capacity of the energy storage. t When it is a negative number, it indicates the storable capacity of the energy storage; and Respectively represent the input capacity of other multi-regional power systems and the output capacity of multi-regional power systems; D t Indicates the load actually absorbed in the multi-regional power system; represents the load reduction due to demand response and load reduction in the multi-regional power system; and It is also subject to line capacity constraints between the multi-regional power system and other multi-regional power systems; 2) System node voltage constraints Where: V min ,V max Respectively represent the maximum and minimum voltage values ​​allowed for each multi-region power system (node); V t,b′ represents the voltage value of any load segment b' of the multi-regional power system in year t; assuming that B' is a time period set, B'={b' / b'=b' x ,x=1,2,3,…},b' x represents the xth time period; 3) Power flow constraints The line capacity between the self-balancing units m and m' is limited, and the active and reactive power that can be carried should be constrained; Where: P m , Q m and |V m | respectively represent the active power, reactive power injection and voltage amplitude of the port of the self-balancing unit m under normal conditions; G mm' and A mm' Respectively represent the real and imaginary parts of the system admittance matrix under normal conditions; θ mm' represents the voltage phase angle difference between the self-balancing units m and m'; 4) Line capacity constraints Where: represents the total capacity of the line between self-balancing units m and m' in year t; represents the total capacity of the line between self-balancing units m and m' in the initial year; represents the increase in line capacity between self-balancing units m and m' in year t; represents the maximum carrying capacity of the line between self-balancing units m and m'; 5) Reliability constraints Considering the actual operation status and economy of the system, the reliability index involved in the model objective function should be limited to a certain value range; 6. The multi-objective collaborative planning method for a power system considering the interconnection of multiple regional power grids according to claim 5 is characterized by: The construction of the two-layer optimization model also includes a lower model, the goal of which is to achieve the optimal configuration of DG, grid architecture, energy storage, and switch elements in the multi-regional power system while minimizing the total investment during the planning period; the model takes the minimum total cost during the planning period as the goal, and comprehensively considers the investment and operating costs of distributed generation, the investment costs and construction costs required for line upgrades and renovations, the investment costs of energy storage units, load reduction costs, and demand response costs, as shown below: The total investment cost of a component is expanded using an auxiliary function: Where: I represents the node set, I = {i / i = i x ,x=1,2,3,…},i x represents the xth node; G represents the generation type set of DG, G={g / g=g x ,x=1,2,3,…}, same as the above model; g represents the g-th generation type of DG; T is the total number of years in the planning period, same as the above model; t represents the t-th year in the planning period; represents the unit capacity investment cost of DG units of power generation type g; represents the newly added DG unit capacity of generation type g in node i in year t; represents the unit capacity operating cost of DG units of generation type g; represents the DG operating capacity of generation type g in node i in year t; h g,i,t represents the annual operation time of the power generation type g in node i in year t; r represents the discount rate; w represents the optional line type when the transmission line is expanded; Represents the unit length line investment and construction cost of line type w; π ij,t represents the number of newly added line loops between nodes i and j in year t; l ij Represents the length between node i and node j; represents a Boolean variable, It indicates that there is expansion between node i and node j in year t, and vice versa; represents the unit capacity operating cost of the line type w; represents the actual operating capacity between nodes i and j in year t; CF S , CO S They represent the unit capacity investment cost and unit operating cost of the energy storage unit respectively; represents the newly added energy storage capacity in node i in year t; represents the total amount of electricity flowing between the system and the energy storage unit in node i in year t; CF Cap Represents the unit capacity investment cost of the parallel capacitor; Q Cap Indicates the rated capacity of each parallel capacitor; represents the number of parallel capacitors installed at node i in year t; C U represents the system outage cost, which is composed of the demand response fee C DR and load reduction cost C cut Composition; B is the peak load segment set, B = {b / b = b x ,x=1,2,3,…},b x represents the xth peak load segment; c DR represents the unit cost of implementing demand response in the system; represents the load reduction through demand response in peak load period b in year t; z b,t is a Boolean variable, z b,t =1, it means that the peak load period b in year t has achieved load reduction, and vice versa; c cut It represents the unit compensation fee paid to users for directly reducing user load; represents the amount of electricity directly reduced by users during peak load period b in year t; z′ b,t is a Boolean variable, z′ b,t =1, it means that the peak load segment b in year t has achieved direct load reduction, and vice versa; N represents the collection of all devices in the system, N = {n / n = n x ,x=1,2,3,…},n x Indicates the nth device; T n Indicates the life span of equipment n; t n Indicates the year when the equipment is put into use, that is, the number of years counted from the initial year of the planning period.

7. The multi-objective collaborative planning method for a power system considering the interconnection of multiple regional power grids according to claim 6 is characterized by: The lower model also includes the following: 1) DG capacity constraints The active power and reactive output of the DG unit are limited by its maximum capacity and minimum capacity. At the same time, the installable DG capacity of node i is also constrained, as shown in the following formula: Where: They represent the upper and lower limits of active output of generation type g in the DG unit respectively; They represent the upper and lower limits of reactive power output of generation type g in the DG unit respectively; represents the capacity of generation type g in node i in the initial year; It represents the total capacity of DG that can be installed on node i; 2) Line capacity constraints The upgrade and expansion of the line needs to consider the constraints of the wiring space; Where: represents the newly added capacity of feeder ij in year t; represents the reactive power flow on feeder ij in year t; represents the power angle of the current from node i to node j in year t; M is a sufficiently large number so that when hour, is 0; when When , the value of M is large enough to ensure that there is enough space to select new lines; Boolean variables Acting as a "switch" for continuous constraints, producing a capacity greater than or equal to the maximum capacity allowed by the line upgrade; 3) Energy storage constraints In actual operation, the charging and discharging capacity and rate of energy storage must meet the following constraints: Where: and are the charging and discharging amounts of the energy storage unit in time period b' respectively; In order to conveniently mark the state of charge (SOC) of the energy storage unit, the t-th year is divided into several load segments. Assume that B' is a set of time periods, B' = {b' / b' = b' x ,x=1,2,3,…},b' x represents the xth time period; η c and η d are the charging and discharging efficiency of the energy storage unit in load segment b'; H i,b′ Indicates the SOC of the energy storage unit during load segment b'; H max and H min are the upper and lower limits of the SOC allowable range respectively; ν b′ represents the charging and discharging rate of the energy storage unit in the load segment b'; ν max Indicates the maximum limit allowed for the charging and discharging rate of the energy storage unit; 4) Power flow constraints Where: P i , Q i and |V i | respectively represent the active power, reactive power injection and voltage amplitude of node i under normal conditions; G ij and A ij Respectively represent the real and imaginary parts of the system admittance matrix under normal conditions; θ ij represents the voltage phase angle difference between node i and node j; 5) Reserve capacity constraints Where: g represents the forced outage rate of DG generation mode g; δ represents the reserve capacity rate; represents the maximum load of node i in year t; 6) Power quality constraints The power quality constraint mainly considers the voltage stability VSI, which is represented by the index ξ; the VSI index ξ of branch I l It can be expressed as: x l ≤V l∈L ij (37) Where: R ij and X ij are the resistance and reactance of branch I respectively; P j and Q j is the active power and reactive power of the receiving end point j of branch I; L ij is the total number of system branches; l The branch corresponding to the maximum value in is called the weakest branch of the system; when the system is stable, the value corresponding to the weakest branch must be less than 1, and the voltage collapse of the system always starts from the weakest branch in the system, ξ l The smaller the voltage, the better the stability; the larger the voltage, the worse the stability; when ξ l When it is close to 1, the system voltage collapses; this constraint needs to ensure that all ξ values ​​are less than a fixed value V<1, thereby ensuring the stability of the system voltage.

8. The multi-objective collaborative planning method for a power system considering the interconnection of multiple regional power grids according to claim 6 is characterized by: The step d comprises the following contents: 1) Algorithm parameter initialization: including harmony memory size HMS, harmony memory selection probability HMCR, pitch fine-tuning probability PAR, pitch fine-tuning bandwidth bw and maximum creation times T max ; 2) Determine the number of tones required in the unit harmony sequence according to the number of DG, energy storage and capacitor nodes to be selected in the system; Each candidate planning scheme corresponds to a harmony in the HSA memory bank, which is encoded in integer form, and its value corresponds to the configuration capacity of DG, energy storage and capacitor in the system; HMS group feasible solution vector is randomly generated; 3) Based on the three-point estimation method, the expected output power of DREG and the probability power flow distribution of the self-balancing unit m under normal operation of each planning scheme are calculated; 4) Combined with the probabilistic power flow distribution during the operation phase, calculate and record the corresponding investment costs of each multi-regional power system and the configuration scheme of DG, energy storage, and capacitor system components; 5) Feedback the expected output power and configuration scheme calculation results of the DREG of the lower model to the upper model as system operation parameters, and calculate the system reliability index and the upper objective function value F(·); 6) Using the linear sorting method to calculate the fitness value J of each individual in the current harmony memory bank, and sorting them in descending order; Where: ζ is the adjustment coefficient; x h Y is the position of the individual after being sorted by the upper layer objective function value; HMS is the number of model constraints; 7) Compare the above harmony individuals with the individuals with the lowest fitness value in the original memory bank. If J(x h,new )>J(x h,hold ), the inferior harmony in the original library is replaced by a better harmony, and the harmony memory library is updated; 8) Determine whether the maximum number of creation times T has been reached max ,If so, output the current optimal harmony as the final planning scheme; Otherwise, perform harmony memory library learning, pitch fine-tuning and pitch random selection operations according to the HSA rule to generate a new harmony library, and return to step 3).

9. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the method according to any one of claims 1 to 8.

10. An electronic device, characterized in that: It comprises a processor and a memory; the memory stores computer-readable instructions, and the processor is used to execute the computer-readable instructions, wherein the computer-readable instructions execute the method described in any one of claims 1 to 8 when executed.

Images