Partitioned joint peak scheduling method, device, medium and product considering day-ahead operation safety domain

Abstract

The present application discloses a method, device, medium and product for joint peak dispatching of zones taking into account the day-ahead operation safety domain, and relates to the field of power dispatching technology. The method includes: solving the overall day-ahead dispatching model after dimensionality reduction to obtain the overall day-ahead dispatching plan; constructing the day-ahead operation safety domain based on the overall day-ahead dispatching plan and the local load of each zone, determining the upward adjustment capacity and downward adjustment capacity under different unit operation points in the day-ahead operation safety domain, and performing cluster screening to obtain the zone adjustable capacity; determining the zone day-ahead dispatching plan of each zone based on the overall day-ahead dispatching plan; constructing and solving the peak demand allocation model to obtain the zone peak demand of each zone; inputting the zone day-ahead dispatching plan and the zone peak demand into the zone plan correction model to obtain the zone day-ahead dispatching correction plan. The present application solves the problem that the day-ahead dispatching plan cannot be quickly corrected in the face of new peak demands.

Description

Technical Field

[0001] The present application relates to the technical field of electric power dispatching, and in particular to a partitioned joint peak dispatching method, device, medium and product taking into account the day-ahead operation safety domain. Background Art

[0002] With the continuous development of the power system and the high proportion of new energy access, the power system is facing complex large-scale dispatching problems under the intensification of bilateral uncertainty between sources and loads. Traditional dispatching methods face problems such as high computational complexity, slow optimization speed, and low computational accuracy when dealing with large-scale dispatching problems. At the same time, the day-ahead dispatch plan obtained by using a large amount of computing resources is obviously not worth the cost when facing new peak demand. Summary of the invention

[0003] The purpose of this application is to provide a partitioned joint peak scheduling method, device, medium and product that takes into account the day-ahead operation safety domain, so as to solve the problem that the day-ahead scheduling plan cannot be quickly corrected when facing new peak demands.

[0004] To achieve the above objectives, this application provides the following solutions.

[0005] In a first aspect, the present application provides a partition joint peak scheduling method considering the day-ahead running safety domain, including:

[0006] Constructing an overall day-ahead scheduling model for the current scheduling area; the overall day-ahead scheduling model includes: an overall day-ahead scheduling objective function and overall constraints;

[0007] A model dimension reduction strategy is adopted to reduce the dimension of the overall day-ahead scheduling model of the current scheduling area, so as to obtain the overall day-ahead scheduling model after dimension reduction of the current scheduling area; the model dimension reduction strategy includes: a time period merging strategy and a strategy for identifying the normally on and normally off thermal power generating units;

[0008] The adaptive binary salp optimization algorithm is used to solve the overall day-ahead scheduling model of the current scheduling area after dimensionality reduction, and the overall day-ahead scheduling plan of the current scheduling area is obtained;

[0009] Based on the overall day-ahead dispatch plan of the current dispatch area and the local load of each subarea of ​​the current dispatch area, the day-ahead operation safety domain of the current dispatch area is constructed, and the upward adjustment capacity and downward adjustment capacity of different unit operating points in the day-ahead operation safety domain are determined;

[0010] The double k-means clustering algorithm is used to cluster and screen the upward adjustment capacity and downward adjustment capacity of each unit under the operating point, and the zone adjustable capacity under the overall day-ahead scheduling plan of the current scheduling area is obtained.

[0011] Determine the day-ahead scheduling plan for each subarea of ​​the current scheduling area based on the overall day-ahead scheduling plan of the current scheduling area;

[0012] Based on the partition adjustable capacity of each partition in the current scheduling area, a peak demand allocation model of the current scheduling area is constructed;

[0013] The Arctic Puffin optimization algorithm is used to solve the peak demand allocation model of the current dispatching area, and the peak demand of each partition in the current dispatching area is obtained;

[0014] The day-ahead scheduling plan and the peak demand of each zone in the current scheduling area are input into the zone plan correction model to obtain the day-ahead scheduling correction plan of the corresponding zone; the zone plan correction model is obtained by training the BiGRU-Attention neural network.

[0015] Optionally, the process of determining the partition plan revision model includes:

[0016] A random function is used to generate a day-ahead scheduling plan and a peak demand for each partition of a plurality of training areas;

[0017] Based on the overall day-ahead scheduling plan of each training scheduling area and the peak demand of each zone, a zone day-ahead scheduling correction model for each training scheduling area is constructed;

[0018] The eel grouper optimization algorithm is used to solve the day-ahead scheduling correction model of each training scheduling area, and the day-ahead scheduling correction plan of each training scheduling area is obtained;

[0019] Based on the forward gated recurrent unit, the backward gated recurrent unit and the attention mechanism, the BiGRU-Attention neural network is constructed;

[0020] The BiGRU-Attention neural network is trained with the partition day-ahead scheduling plan and the partition peak demand of each partition in each training scheduling area as input and the partition day-ahead scheduling correction plan of the corresponding partition as output to obtain the partition plan correction model.

[0021] Optionally, the overall day-ahead scheduling objective function includes:

[0022] ;

[0023] ;

[0024] in, is the total cost of day-ahead scheduling; The cost of thermal power; Cost of curtailed electricity; The operating cost of energy storage; for demand response costs; The cost of coal consumption; is the start-stop cost; is the spinning reserve cost; The number of thermal power units in the dispatching area; is the scheduling period; is the start / stop state variable of the i-th thermal power unit at time t; , and All are coal consumption coefficients for thermal power generation; is the actual output of the i-th thermal power unit at time t; is the single start-up and shutdown cost of the i-th thermal power unit at time t; is the spinning reserve capacity cost coefficient of the i-th thermal power unit at time t; is the positive spinning reserve capacity of the i-th generator set at time t; is the negative spinning reserve capacity of the i-th thermal power unit at time t; The number of wind turbines in the dispatching area; is the power abandonment cost coefficient of the i-th wind turbine at the t-th moment; is the predicted output of the i-th wind turbine at the t-th moment; is the actual output of the i-th wind turbine at time t; The number of photovoltaic units in the dispatching area; is the power abandonment cost coefficient of the i-th photovoltaic unit at the t-th moment; is the predicted output of the i-th photovoltaic unit at the t-th moment; is the actual output of the i-th photovoltaic unit at the t moment; The number of electrochemical energy storage systems in the dispatching area; is the unit power operation cost of the i-th electrochemical energy storage system at the t-th moment; is the discharge power of the i-th electrochemical energy storage system at the t-th moment; is the charging power of the i-th electrochemical energy storage system at the t-th moment; The number of demand responses in the dispatch area; is the cost coefficient of the ith demand response at time t; is the i-th actual demand response at time t.

[0025] Optionally, the overall constraints include: wind and solar power safe operation range constraints, thermal power unit technical output upper and lower limits and group ramp rate constraints, energy storage system operation constraints, power balance constraints, line flow safety constraints and tie line constraints, system positive and negative spinning reserve constraints and demand response model constraints;

[0026] The wind and solar power safe operation range constraints include:

[0027] ;

[0028] in, is the minimum output of the i-th wind turbine at the t-th moment; is the maximum output of the i-th wind turbine at time t; is the minimum output of the i-th photovoltaic unit at the t-th moment; is the maximum output of the i-th photovoltaic unit at the t moment;

[0029] The technical output upper and lower limits of the thermal power unit and the group climbing rate constraints include:

[0030] ;

[0031] ;

[0032] in, is the minimum output of the i-th thermal power unit at the t-th moment; is the maximum output of the i-th thermal power unit at time t; is the actual output of the i-th thermal power unit at time t+1; is the ramp power limit of the i-th thermal power unit; is the minimum output of the i-th thermal power unit at the t+1th moment; is the start / stop state variable of the i-th thermal power unit at the t+1th moment; is the maximum output of the i-th thermal power unit at time t+1; is the ramp-down power limit of the i-th thermal power unit;

[0033] The energy storage system operation constraints include:

[0034] ;

[0035] in, is the power lower limit of the i-th electrochemical energy storage system at the t-th moment; is the power upper limit of the i-th electrochemical energy storage system at the t-th moment; is the charge state of the i-th electrochemical energy storage system at the time t+1; is the charge state of the i-th electrochemical energy storage system at the t-th moment; is the charging efficiency of the i-th electrochemical energy storage system at the t-th moment; is the discharge efficiency of the i-th electrochemical energy storage system at the t-th moment; is the tth scheduling period; is the rated energy of the i-th electrochemical energy storage system; is the low value of the state of charge of the i-th electrochemical energy storage system at the t-th moment; is the high value of the state of charge of the i-th electrochemical energy storage system at the t-th moment;

[0036] The power balance constraint includes:

[0037] ;

[0038] in, is the i-th local load at time t; is the outgoing load of the jth tie line at time t; The number of tie lines for external load transmission;

[0039] The line flow safety constraints and tie line constraints include:

[0040] ;

[0041] ;

[0042] in, is the upper power limit of the lth line; is the actual power of line No. l at time t; is the upper limit of the power of the jth tie line; is the actual power of the jth tie line;

[0043] The positive and negative spinning reserve constraints of the system include:

[0044] ;

[0045] ;

[0046] in, is the total positive spinning reserve capacity at time t; is the total negative spinning reserve capacity at time t; is the upper reserve capacity factor of the wind turbine; is the upper reserve capacity factor of the PV unit; is the upper reserve capacity factor of the local load; is the lower reserve capacity factor of the wind turbine; is the lower reserve capacity factor of the PV unit; is the lower reserve capacity factor of the local load;

[0047] The demand response model constraints include:

[0048] ;

[0049] in, is the lower limit of the ith demand response; is the peak value of the i-th local load before the day; is the maximum coefficient of the i-th demand response.

[0050] Optionally, the recently-running security domain includes:

[0051] ;

[0052] in, It is the set of operating points of all units in the partition; It is the key decision vector corresponding to all the unit operation points in the partition within the dispatch cycle; is the discrete variable of the partitioned wind-solar-thermal-storage system; is the vector of the remaining variables of the system; is the power flow equation constraint; The unit combination constraint set includes: the wind and solar power safe operation range constraint, the upper and lower limits of the thermal power unit technical output and the group climbing rate constraint, the energy storage system operation constraint and the demand response model constraint; is the security constraint set;

[0053] The expression of the power flow equality constraint is:

[0054] ;

[0055] in, is the number of thermal power units in the xth zone; is the number of wind turbines in the xth zone; is the number of PV panels in the xth zone; is the number of electrochemical energy storage systems in the x-th partition; is the number of local loads in the xth partition; is the number of tie lines in the xth zone;

[0056] The expression of the safety constraint set is:

[0057] ;

[0058] in, is the actual power of the lth line in the xth partition at the tth moment; is the power upper limit of the lth line in the xth partition; is the actual power of the jth tie line in the xth partition at the tth moment; is the upper power limit of the j-th tie line in the x-th partition at the t-th moment; is the positive spinning reserve capacity of the xth partition at time t; is the negative spinning reserve capacity of the xth partition at time t.

[0059] Optionally, the calculation formula for the upward adjustment capability includes:

[0060] ;

[0061] The calculation formula of the downward adjustment capability includes:

[0062] ;

[0063] in, is the upward adjustment capacity of the x-th partition at time t; is the downward adjustment capacity of the x-th partition at the t-th moment.

[0064] Optionally, the peak demand allocation model includes:

[0065] ;

[0066] ;

[0067] ;

[0068] in, is the load demand of the newly added j-th tie line; is the number of partitions; Support power for the peak demand of the xth partition at time t; is the upper limit of the power of the No. 1 tie line in the xth zone; is the power transfer distribution factor between the xth partition and the tie line No. 1; is the power transfer distribution factor between the lth tie line and the jth tie line; is the power flow of the lth line at the tth moment under the day-ahead scheduling plan; is the power upper limit of the jth tie line at the tth moment; is the lower limit of the regulation capacity of the xth partition at the tth moment; is a binary state variable characterizing the regulation capacity mode of the x-th partition at time t; is the upper limit of the regulation capacity of the x-th partition at the t-th moment; Set a threshold for the number of mode switches for the xth partition; is the target value that characterizes the equilibrium peak.

[0069] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-mentioned partitioned joint peak scheduling methods taking into account the day-ahead running security domain.

[0070] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-mentioned partitioned joint peak scheduling methods considering the day-ahead running security domain.

[0071] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned partitioned joint peak scheduling methods considering the day-ahead running security domain.

[0072] According to the specific embodiments provided by the present application, the present application discloses the following technical effects: the present application discloses a partitioned joint peak scheduling method, device, medium and product considering the day-ahead operation safety domain, adopts a model dimensionality reduction strategy to reduce the dimension of decision variables, reduce computational complexity, and realizes rapid solution of optimization problems involving many binary variables through an adaptive binary salp optimization algorithm, and uses an adaptive step size to improve the global search capability of the algorithm; the day-ahead operation safety domain can describe the range of the unit's operating points, and quantify the partitioned adjustable capabilities of the unit's operating points, and the dual k-means clustering algorithm can be used to screen the adjustment capabilities, thereby obtaining the partitioned adjustable capabilities, which is beneficial to the partitioning Visualization of internal adjustable capacity; the power transfer distribution factor is used to provide an accurate description of peak demand transmission, and the Arctic Puffin optimization algorithm is used to solve the peak demand allocation model, which ensures the balanced distribution of peak demand and achieves rapid convergence of the allocation process to improve allocation efficiency; the correction factor is introduced to correct the day-ahead scheduling plan and ensure the feasibility of the correction plan, reducing the difficulty of implementing the day-ahead scheduling correction plan. The eel grouper optimization algorithm can quickly find the optimal solution in complex scheduling correction problems and improve the optimality of the day-ahead scheduling correction plan. The BiGRU-Attention neural network model is further used to realize the rapid acquisition of the partitioned day-ahead scheduling correction plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0074] Figure 1A flowchart of a partitioned joint peak scheduling method considering the day-ahead running security domain is provided for one embodiment of the present application.

[0075] Figure 2 Schematic diagram of the partitioned joint peak scheduling architecture considering the day-ahead operation safety domain.

[0076] Figure 3 This is an overview diagram of partitioned joint peak scheduling considering the day-ahead operation safety domain.

[0077] Figure 4 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0079] The purpose of this application is to provide a partitioned joint peak scheduling method, device, medium and product that takes into account the day-ahead operation safety domain, aiming to solve the problem that the day-ahead scheduling plan cannot be quickly corrected when facing new peak demands.

[0080] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0081] In an exemplary embodiment, Figure 1-Figure 3 As shown, the partition joint peak scheduling method considering the day-ahead running safety domain in this embodiment includes the following steps.

[0082] Step 1: Construct the overall day-ahead scheduling model for the current scheduling area.

[0083] Among them, the overall day-ahead scheduling model includes: the overall day-ahead scheduling objective function and the overall constraints.

[0084] Specifically, based on the day-ahead forecast data of wind power, photovoltaic power and electric load, an output dispatch plan for thermal power units, wind power and photovoltaic power units and electrochemical energy storage systems in the day-ahead wind, solar and thermal storage system is formulated, and the demand response of electric load is further considered to establish an overall day-ahead dispatch model.

[0085] As an optional implementation, the overall day-ahead scheduling objective function includes:

[0086] .

[0087] .

[0088] in, is the total cost of day-ahead scheduling; The cost of thermal power; Cost of curtailed electricity; The operating cost of energy storage; for demand response costs; The cost of coal consumption; is the start-stop cost; is the spinning reserve cost; The number of thermal power units in the dispatching area; is the scheduling period; is the start / stop state variable of the i-th thermal power unit at time t; , and All are coal consumption coefficients for thermal power generation; is the actual output of the i-th thermal power unit at time t; is the single start-up and shutdown cost of the i-th thermal power unit at time t; is the spinning reserve capacity cost coefficient of the i-th thermal power unit at time t; is the positive spinning reserve capacity of the i-th generator set at time t; is the negative spinning reserve capacity of the i-th thermal power unit at time t; The number of wind turbines in the dispatching area; is the power abandonment cost coefficient of the i-th wind turbine at the t-th moment; is the predicted output of the i-th wind turbine at the t-th moment; is the actual output of the i-th wind turbine at time t; The number of photovoltaic units in the dispatching area; is the power abandonment cost coefficient of the i-th photovoltaic unit at the t-th moment; is the predicted output of the i-th photovoltaic unit at the t-th moment; is the actual output of the i-th photovoltaic unit at the t moment; The number of electrochemical energy storage systems in the dispatching area; is the unit power operation cost of the i-th electrochemical energy storage system at the t-th moment; is the discharge power of the i-th electrochemical energy storage system at the t-th moment; is the charging power of the i-th electrochemical energy storage system at the t-th moment; The number of demand responses in the dispatch area; is the cost coefficient of the ith demand response at time t; is the i-th actual demand response at time t.

[0089] As an optional implementation method, the overall constraints include: wind and solar safe operation range constraints, upper and lower limits of thermal power unit technical output and group ramp rate constraints, energy storage system operation constraints, power balance constraints, line flow safety constraints and interconnection line constraints, system positive and negative rotating reserve constraints and demand response model constraints.

[0090] The safe operation range constraints of wind and solar power include:

[0091] .

[0092] in, is the minimum output of the i-th wind turbine at the t-th moment; is the maximum output of the i-th wind turbine at time t; is the minimum output of the i-th photovoltaic unit at the t-th moment; is the maximum output of the i-th PV unit at the t-th moment.

[0093] The upper and lower limits of thermal power unit technical output and unit ramp rate constraints include:

[0094] .

[0095] .

[0096] in, is the minimum output of the i-th thermal power unit at the t-th moment; is the maximum output of the i-th thermal power unit at time t; is the actual output of the i-th thermal power unit at time t+1; is the ramp power limit of the i-th thermal power unit; is the minimum output of the i-th thermal power unit at the t+1th moment; is the start / stop state variable of the i-th thermal power unit at the t+1th moment; is the maximum output of the i-th thermal power unit at time t+1; is the ramp-down power limit of the i-th thermal power unit.

[0097] Energy storage system operation constraints, including:

[0098] .

[0099] in, is the power lower limit of the i-th electrochemical energy storage system at the t-th moment; is the power upper limit of the i-th electrochemical energy storage system at the t-th moment; is the charge state of the i-th electrochemical energy storage system at the time t+1; is the charge state of the i-th electrochemical energy storage system at the t-th moment; is the charging efficiency of the i-th electrochemical energy storage system at the t-th moment; is the discharge efficiency of the i-th electrochemical energy storage system at the t-th moment; is the tth scheduling period; is the rated energy of the i-th electrochemical energy storage system; is the low value of the state of charge of the i-th electrochemical energy storage system at the t-th moment; is the high value of the state of charge of the i-th electrochemical energy storage system at the t-th moment.

[0100] Power balance constraints, including:

[0101] .

[0102] in, is the i-th local load at time t; is the outgoing load of the jth tie line at time t; The number of interconnection lines for external load transmission.

[0103] Line flow safety constraints and tie line constraints, including:

[0104] .

[0105] .

[0106] in, is the upper power limit of the lth line; is the actual power of line No. l at time t; is the upper limit of the power of the jth tie line; is the actual power of the jth tie line.

[0107] System positive and negative spinning reserve constraints, including:

[0108] .

[0109] .

[0110] in, is the total positive spinning reserve capacity at time t; is the total negative spinning reserve capacity at time t; is the upper reserve capacity factor of the wind turbine; is the upper reserve capacity factor of the PV unit; is the upper reserve capacity factor of the local load; is the lower reserve capacity factor of the wind turbine; is the lower reserve capacity factor of the PV unit; is the lower reserve capacity factor of the local load.

[0111] Demand response model constraints, including:

[0112] .

[0113] in, is the lower limit of the ith demand response; is the peak value of the i-th local load before the day; is the maximum coefficient of the i-th demand response.

[0114] Step 2: Adopt the model dimension reduction strategy to reduce the dimension of the overall day-ahead scheduling model of the current scheduling area, and obtain the overall day-ahead scheduling model after dimension reduction in the current scheduling area.

[0115] Among them, the model dimension reduction strategies include: time period merging strategy and often-on and often-off generating units identification strategy.

[0116] Specifically, the time period merging strategy merges the calculation cycles according to the changing trend of the electric load in different periods of the calculation cycle. The net load change rate in the dispatching area is used as the basis for division:

[0117] .

[0118] in, is the net load at time t; is the net load change rate at time t.

[0119] If the net load change rate is less than the set threshold, the time periods will be merged. To ensure the consistency of electric energy after the time periods are merged, the net load and the corresponding time periods will be equivalently calculated:

[0120] .

[0121] .

[0122] in, is the merge threshold; is the combined equivalent net load at time t; is the net load at time t+1; is the equivalent period after the merger; It is the t+1th scheduling period.

[0123] The identification strategy of normally open and normally shut-down thermal power units divides thermal power units into normally open units, normally shut-down units, and start-stop units according to the preset unit operation information. In practice, thermal power units are usually in full-time start-up or shutdown state, and only some units need to be started or shut down. By using this feature, the integer variables that need to be branched in the model can be greatly reduced, thereby greatly reducing the state combination space of the system.

[0124] Step 3: Use the adaptive binary salp optimization algorithm to solve the overall day-ahead scheduling model after dimensionality reduction of the current scheduling area to obtain the overall day-ahead scheduling plan of the current scheduling area.

[0125] Specifically, the Salp Swarm Algorithm (SSA) initializes a set of random positions for each group, regards the best position of all groups as the target group, and other groups chase it, so that the position update is divided into two stages, namely the leadership stage and the following stage. During the optimization process, followers may not follow the leader and fall into the local optimal solution. The mutation operator is further introduced to enrich the population diversity and the search ability of the algorithm. For binary variables, the Arctan transfer function is used to explore and utilize the binary search space.

[0126] .

[0127] in, is a random number distributed between two elements; is a random function; is the maximum value of the elements in the a-th group; is the minimum value of the elements in the a-th group; is the location information of group a; Adaptive step size can prevent premature convergence and thus produce smooth convergence; and All are constant coefficients of step length; is the maximum number of iterations; is the current iteration number; is the binary variable decision variable for the bth element in the ath group; is the Arctan transfer function.

[0128] The specific solution process of the Adaptive Binary Salp Swarm Algorithm (ABSSA) is as follows: First, the decision variables are initialized based on the overall day-ahead scheduling model boundary to generate a population; second, the position of the continuous decision variables is updated using the leading and following phases of SSA, and the Arctan transfer function is used to update the binary variables; then, the step size is calculated. , further calculate the quasi-oppositional population , and according to the change of the target fitness of the a-th group, choose whether to update the position of the a-th group. Finally, update the best position of the group and increase the iteration count. When the maximum iteration is reached, the loop is terminated.

[0129] Step 4: Based on the overall day-ahead dispatch plan of the current dispatch area and the local load of each partition of the current dispatch area, construct the day-ahead operation safety domain of the current dispatch area, and determine the upward adjustment capacity and downward adjustment capacity at different unit operating points in the day-ahead operation safety domain.

[0130] Specifically, based on the start and stop plan of the generator sets within the zone and the local load of the zone, the day-ahead operation safety domain is expressed as: taking into account the power supply structure characteristics and power demand distribution characteristics of the zone, and under the start and stop states of the units arranged in the overall day-ahead scheduling plan, the set of unit operating points that meet safe operation and power balance within the scheduling cycle.

[0131] As an optional implementation, the security domain is currently running, including:

[0132] .

[0133] in, It is the set of operating points of all units in the partition; It is the key decision vector corresponding to all the unit operation points in the partition within the dispatch cycle; is the discrete variable of the partitioned wind-solar-thermal-storage system; is the vector of the remaining variables of the system; is the power flow equation constraint; The unit combination constraint set includes: wind and solar power safe operation range constraints, thermal power unit technical output upper and lower limits and group ramp rate constraints, energy storage system operation constraints and demand response model constraints; is a set of security constraints.

[0134] The expression of the power flow equality constraint is:

[0135] .

[0136] in, is the number of thermal power units in the xth zone; is the number of wind turbines in the xth zone; is the number of PV panels in the xth zone; is the number of electrochemical energy storage systems in the x-th partition; is the number of local loads in the xth partition; is the number of tie lines in the xth partition.

[0137] The expression of the safety constraint set is:

[0138] .

[0139] in, is the actual power of the lth line in the xth partition at the tth moment. is the power upper limit of the lth line in the xth partition. is the actual power of the jth tie line in the xth partition at the tth moment. is the upper power limit of the j-th tie line in the x-th partition at the t-th moment. is the positive spinning reserve capacity of the xth partition at time t. is the negative spinning reserve capacity of the xth partition at time t.

[0140] As an optional implementation, the calculation formula for upward adjustment capability includes:

[0141] .

[0142] The calculation formula for downward adjustment capacity includes:

[0143] .

[0144] in, is the upward adjustment capacity of the x-th partition at time t; is the downward adjustment capacity of the x-th partition at the t-th moment.

[0145] Step 5: Use the double k-means clustering algorithm to cluster and screen the upward adjustment capacity and downward adjustment capacity of each unit operating point, and obtain the partitioned adjustable capacity under the overall day-ahead scheduling plan of the current scheduling area.

[0146] Specifically, under different unit operation points in the day-ahead operation safety domain, the adjustment capabilities are inconsistent. In order to screen out the maximum upward adjustment capability and downward adjustable capability, a double k-means clustering algorithm is used for clustering screening, including the following steps.

[0147] In the first step, the Monte Carlo sampling method is used to generate a large number of unit operating points in the day-ahead operation safety region, and the upward adjustment capacity and downward adjustment capacity are calculated according to the calculation formula of the upward adjustment capacity and the calculation formula of the downward adjustment capacity. The adjustment capacity of the kth unit operating point is .

[0148] In the second step, the k-means clustering algorithm is used to take the upward adjustment ability and the downward adjustment ability in all time periods as the clustering point sets, and sub-clusters are generated after clustering; further, in the sub-clusters, the upward adjustment ability and the downward adjustment ability in all time periods are used as the clustering point sets, and upward adjustment ability sub-sub-clusters and downward adjustment ability sub-sub-clusters are generated after clustering.

[0149] In the third step, the cluster point set in the sub-sub-cluster is averaged and summed to obtain the sub-sub-cluster target value. The sub-sub-cluster target value with the largest upward adjustment ability and the sub-sub-cluster target value with the largest downward adjustment ability in the sub-cluster are further selected to represent the sub-cluster.

[0150] In the fourth step, the representative values ​​of all sub-clusters are compared. Since the maximum upward adjustment ability and the maximum downward adjustment ability cannot be achieved at the same time, the sub-cluster representative with the maximum upward adjustment ability and the corresponding sub-cluster downward adjustment ability, and the sub-cluster representative with the maximum downward adjustment ability and the corresponding sub-cluster upward adjustment ability are selected respectively, and two solutions are obtained:

[0151] .

[0152] in, is the maximum upward adjustment capacity of the xth partition at time t; is the downward adjustment capacity corresponding to the maximum upward adjustment capacity of the x-th partition at time t; is the maximum downward adjustment capacity of the x-th partition at time t; is the upward adjustment capacity corresponding to the maximum downward adjustment capacity of the x-th partition at time t; and are the matching coefficients of the x-th partition at the t-th moment.

[0153] Step 6: Determine the day-ahead scheduling plan for each partition of the current scheduling area based on the overall day-ahead scheduling plan for the current scheduling area.

[0154] Step 7: Based on the partition adjustable capacity of each partition in the current scheduling area, a peak demand allocation model of the current scheduling area is constructed.

[0155] Specifically, according to the partition structure, the partition is regarded as a generalized node, and only the input and output characteristics are concerned. The power lines connecting the partitions in the original area are regarded as the newly added interval interconnection lines of the partitions. The interval interconnection lines between partitions are used to establish the power transfer distribution factor to reflect the impact of the input and output of the partition generalized nodes on the flow of the interval interconnection lines:

[0156] .

[0157] in, is the power transfer distribution factor between the i-th node and the k-th tie line branch; is the reactance value between the mth node and the ith node; is the reactance value between the nth node and the i-th node; and is the parameter in the reactance matrix of the power grid branch; is the reactance value of the kth tie line branch.

[0158] As an optional implementation, the peak demand comes from the transmission load of the tie line. Considering the regulation capacity of the partition, the sudden increase of the peak demand of the tie line will be divided into each partition in the dispatching area. The peak demand allocation model includes: peak demand constraints, linear constraints and balanced peak objective function.

[0159] (1) The allocation of peak demand should meet peak demand constraints such as power balance constraints, interconnection line safety flow constraints, external interconnection line constraints, and regulation capacity constraints. The peak demand constraints include:

[0160] .

[0161] (2) Using binary variables, the regulation capacity constraint is linearized. At the same time, the number of regulation capacity mode switching times should be less than the set threshold. The linearization constraint is:

[0162] .

[0163] (3) In order to take into account the adjustable capacity of each partition and make each area participate in the peak link in a balanced manner, the following balanced peak objective function is established:

[0164] .

[0165] in, is the load demand of the newly added j-th tie line; is the number of partitions; Support power for the peak demand of the xth partition at time t; is the upper limit of the power of the No. 1 tie line in the xth zone; is the power transfer distribution factor between the xth partition and the tie line No. 1; is the power transfer distribution factor between the lth tie line and the jth tie line; is the power flow of the lth line at the tth moment under the day-ahead scheduling plan; is the power upper limit of the jth tie line at the tth moment; is the lower limit of the regulation capacity of the xth partition at the tth moment; is a binary state variable characterizing the regulation capacity mode of the x-th partition at time t; is the upper limit of the regulation capacity of the x-th partition at the t-th moment; Set a threshold for the number of mode switches for the xth partition; is the target value that characterizes the equilibrium peak.

[0166] Step 8: Use the Arctic Puffin optimization algorithm to solve the peak demand allocation model of the current scheduling area and obtain the partition peak demand of each partition in the current scheduling area.

[0167] Specifically, the Arctic Puffin Optimization (APO) algorithm consists of aerial flight (exploration) and underwater foraging (exploitation) stages. In the exploration stage, the flight and speed factor mechanisms are introduced to enhance the algorithm's ability to jump out of the local optimum and improve the convergence speed. In the development stage, strategies such as collaborative and adaptive change factors are adopted to ensure that the algorithm can effectively utilize the current optimal solution and guide the search direction. The exploration stage is divided into; aerial search strategy and diving predation strategy, and the development stage is divided into collective foraging strategy, enhanced search strategy and predator avoidance strategy. The specific iterative equation is as follows:

[0168] .

[0169] .

[0170] in, It is the candidate solution after the strategy update in the first stage; is the hth candidate solution in the population; is randomly selected from Other candidate solutions of ; It is a random number generated during flight; is a random parameter about the standard normal distribution; It is the candidate solution after the second-stage strategy update; is the speed coefficient; The candidate solution after the set foraging strategy is updated; , , are randomly selected mutually distinct candidate solutions; is the cooperation factor; is the adaptive factor; is a uniformly distributed random number between 0 and 1.

[0171] The specific implementation process of the APO algorithm is as follows: first, the population is initialized according to the boundary of the partition decision variable. Secondly, the conversion factor related to the number of iterations is used to determine whether to enter the aerial flight stage or the underwater foraging stage. Furthermore, the aerial search strategy and the diving predation strategy are executed in sequence in the aerial flight stage, and the collective foraging strategy, the enhanced search strategy and the predator avoidance strategy are executed in sequence in the underwater foraging stage. The former strategy with the best fitness value among all strategies in the stage is selected. individuals to form a new population; finally, iterate continuously until the upper limit of the number of iterations is reached.

[0172] Step 9: Input the day-ahead scheduling plan and the peak demand of each zone in the current scheduling area into the zone plan correction model to obtain the day-ahead scheduling correction plan of the corresponding zone.

[0173] Among them, the partition plan correction model is obtained by training the BiGRU-Attention neural network.

[0174] As an optional implementation, the process of determining the partition plan revision model includes the following steps.

[0175] Step 91: Use a random function to generate a partition day-ahead scheduling plan and a partition peak demand for each partition of a plurality of training areas.

[0176] Step 92: Based on the overall day-ahead scheduling plan of each training scheduling area and the zone peak demand of each zone, a zone day-ahead scheduling correction model for each training scheduling area is constructed.

[0177] Specifically, based on the overall day-ahead dispatch plan, in order to meet the peak demand of the sub-area, the output plan of the generator set needs to be readjusted within the sub-area, and the marginal cost of the generator set under the overall day-ahead dispatch plan is regarded as the correction factor. The lower the marginal cost of the generator set, the smaller its correction factor. By presetting the correction factor for the generator set, the day-ahead dispatch plan of the sub-area can be quickly corrected. The day-ahead dispatch correction model of the sub-area includes: deviation objective function and internal constraints.

[0178] (1) Considering the thermal power units, demand response and electrochemical energy storage systems that provide adjustable capacity, the deviation of the overall day-ahead dispatch plan is minimized, and the deviation objective function is constructed as follows:

[0179] .

[0180] in, is the deviation; is the correction factor of the i-th thermal power unit at the t-th moment; is the output of the i-th thermal power unit under the day-ahead dispatch plan; is the number of local loads in the xth partition; is the correction factor of the i-th demand response at time t; is the i-th demand response at time t under the day-ahead dispatch plan; is the correction factor of the i-th electrochemical energy storage system at the t-th moment; is the discharge power of the i-th electrochemical energy storage system at the t-th time under the day-ahead scheduling plan; is the charging power of the electrochemical energy storage system at time t under the i-th day-ahead scheduling plan.

[0181] (2) Internal constraints should be met within the zone. Internal constraints include: the following formula as well as the upper and lower limits of the thermal power unit's technical output and the unit's ramp rate constraints, energy storage system operation constraints, and demand response model constraints.

[0182] .

[0183] in, is the output of the i-th wind turbine at time t under the day-ahead scheduling plan; is the output of the i-th PV unit at time t under the day-ahead scheduling plan; is the increment of the outgoing load of the jth tie line at time t; is the power transfer distribution factor between the lth line and the i-th node in the x-th partition; is the power injection amount of the i-th node at the t-th time under the day-ahead scheduling plan; is the power transfer distribution factor between the lth line and the jth node in the xth partition; is the power transfer distribution factor between the lth line and the pth node in the xth partition.

[0184] Step 93: Use the eel grouper optimization algorithm to solve the partitioned day-ahead scheduling correction model of each training scheduling area respectively, and obtain the partitioned day-ahead scheduling correction plan of each training scheduling area.

[0185] Specifically, in the face of large-scale optimization problems, the Eel and Grouper Optimizer (EGO) algorithm has a strong ability to balance development and exploration capabilities. The EGO algorithm uses the symbiotic foraging relationship between eels and groupers in the marine ecosystem and divides the optimization process into three stages: groupers tracking prey (tracking stage), groupers sending signals to eels (signaling stage), and cooperative hunting (hunting stage). The specific mathematical model is as follows:

[0186] .

[0187] .

[0188] .

[0189] in; is the position information of individual s in the u+1th iteration; is the random position of the sth individual in the uth iteration; and All are random coefficients; is the position information of individual No. s at the u-th iteration; is the fitness function for calculating the position target value; is the best position of individual s at the u+1th iteration; is the best position of individual s at the u-th iteration; is the location information of the s-th subordinate individual in the u-th iteration; is a random number between 0 and 1; is the current iteration number; is the maximum number of iterations; and is the position distance between the subordinate individuals and the individuals; is a random number between 0 and 1.

[0190] The specific implementation process of the EGO algorithm is as follows: first, the population of decision variables is initialized; second, the tracking phase is carried out to update the optimal position according to the position target value; third, the signaling phase is carried out to update the position information of the subordinate individuals; fourth, according to the random function, the position distance between the subordinate individuals and the individuals is used to update the individual position update information of the next iteration; finally, iterate continuously until the maximum number of iterations is reached.

[0191] Step 94: Construct a BiGRU-Attention neural network based on the forward gated recurrent unit, the backward gated recurrent unit and the attention mechanism.

[0192] Step 95: Taking the day-ahead scheduling plan and the peak demand of each partition in each training scheduling area as input and the day-ahead scheduling correction plan of the corresponding partition as output, the BiGRU-Attention neural network is trained to obtain a partition plan correction model.

[0193] Specifically, when the day-ahead scheduling plan and the peak demand of the partition are known, the partition plan correction model constructed based on the BiGRU-Attention neural network can be used to quickly obtain the partition day-ahead scheduling correction plan. Specifically, it includes: the first step is to use a random function to generate a large number of partition day-ahead scheduling plans and partition peak demands, and the corresponding partition day-ahead scheduling correction plan is obtained after solving the eel grouper optimization algorithm. Further, a massive neural network data set is constructed based on the massive partition day-ahead scheduling plans, partition peak demands and partition day-ahead scheduling correction plans. The second step is to construct a BiGRU-Attention neural network. BiGRU is a network model composed of a forward gated recurrent unit and a backward gated recurrent unit. The deep extraction of features is achieved through the forward and reverse calculation of the time series. The attention mechanism can prevent the features containing important information from disappearing as the step length increases by assigning different weights to the input features. The third step is to use a massive neural network data set consisting of the partition day-ahead scheduling plan, partition peak demand and partition day-ahead scheduling correction plan of each partition in each training scheduling area to train the BiGRU-Attention neural network. After the training is completed, the partition plan correction model can be obtained.

[0194] In an exemplary embodiment, a computer device is provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a partitioned joint peak scheduling method that considers a day-ahead running safety domain.

[0195] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, a partitioned joint peak scheduling method considering a day-ahead running security domain is implemented.

[0196] In an exemplary embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements a partitioned joint peak scheduling method considering a day-ahead running security domain.

[0197] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a partitioned joint peak scheduling method considering the day-ahead running security domain is implemented.

[0198] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0199] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0200] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0201] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0202] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0203] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A partitioned joint peak scheduling method considering the day-ahead operation safety domain, characterized in that: The partition joint peak scheduling method considering the day-ahead running safety domain includes: Constructing an overall day-ahead scheduling model for the current scheduling area; the overall day-ahead scheduling model includes: an overall day-ahead scheduling objective function and overall constraints; A model dimension reduction strategy is adopted to reduce the dimension of the overall day-ahead scheduling model of the current scheduling area, so as to obtain the overall day-ahead scheduling model after dimension reduction of the current scheduling area; the model dimension reduction strategy includes: a time period merging strategy and a strategy for identifying the normally on and normally off thermal power generating units; The adaptive binary salp optimization algorithm is used to solve the overall day-ahead scheduling model of the current scheduling area after dimensionality reduction, and the overall day-ahead scheduling plan of the current scheduling area is obtained; Based on the overall day-ahead dispatch plan of the current dispatch area and the local load of each subarea of ​​the current dispatch area, the day-ahead operation safety domain of the current dispatch area is constructed, and the upward adjustment capacity and downward adjustment capacity of different unit operating points in the day-ahead operation safety domain are determined; The double k-means clustering algorithm is used to cluster and screen the upward adjustment capacity and downward adjustment capacity of each unit under the operating point, and the zone adjustable capacity under the overall day-ahead scheduling plan of the current scheduling area is obtained. Determine the day-ahead scheduling plan for each subarea of ​​the current scheduling area based on the overall day-ahead scheduling plan of the current scheduling area; Based on the partition adjustable capacity of each partition in the current scheduling area, a peak demand allocation model of the current scheduling area is constructed; The Arctic Puffin optimization algorithm is used to solve the peak demand allocation model of the current dispatching area, and the peak demand of each partition in the current dispatching area is obtained; The day-ahead scheduling plan and the peak demand of each zone in the current scheduling area are input into the zone plan correction model to obtain the day-ahead scheduling correction plan of the corresponding zone; the zone plan correction model is obtained by training the BiGRU-Attention neural network.

2. The partition joint peak scheduling method considering the day-ahead operation safety domain according to claim 1 is characterized in that: The process of determining the partition plan revision model includes: A random function is used to generate a day-ahead scheduling plan and a peak demand for each partition of a plurality of training areas; Based on the overall day-ahead scheduling plan of each training scheduling area and the peak demand of each zone, a zone day-ahead scheduling correction model for each training scheduling area is constructed; The eel grouper optimization algorithm is used to solve the day-ahead scheduling correction model of each training scheduling area, and the day-ahead scheduling correction plan of each training scheduling area is obtained; Based on the forward gated recurrent unit, the backward gated recurrent unit and the attention mechanism, the BiGRU-Attention neural network is constructed; The BiGRU-Attention neural network is trained with the partition day-ahead scheduling plan and the partition peak demand of each partition in each training scheduling area as input and the partition day-ahead scheduling correction plan of the corresponding partition as output to obtain the partition plan correction model.

3. The partition joint peak scheduling method considering the day-ahead operation safety domain according to claim 1 is characterized in that: The overall day-ahead scheduling objective function includes: ； ； in, is the total cost of day-ahead scheduling; The cost of thermal power; Cost of curtailed electricity; The operating cost of energy storage; for demand response costs; The cost of coal consumption; is the start-stop cost; is the spinning reserve cost; is the number of thermal power units in the dispatching area; is the scheduling period; is the start / stop state variable of the i-th thermal power unit at time t; , and All are coal consumption coefficients for thermal power generation; is the actual output of the i-th thermal power unit at time t; is the single start-up and shutdown cost of the i-th thermal power unit at time t; is the spinning reserve capacity cost coefficient of the i-th thermal power unit at time t; is the positive spinning reserve capacity of the i-th generator set at time t; is the negative spinning reserve capacity of the i-th thermal power unit at time t; The number of wind turbines in the dispatching area; is the power abandonment cost coefficient of the i-th wind turbine at the t-th moment; is the predicted output of the i-th wind turbine at the t-th moment; is the actual output of the i-th wind turbine at time t; The number of photovoltaic units in the dispatching area; is the power abandonment cost coefficient of the i-th photovoltaic unit at the t-th moment; is the predicted output of the i-th photovoltaic unit at the t-th moment; is the actual output of the i-th photovoltaic unit at the t moment; The number of electrochemical energy storage systems in the dispatching area; is the unit power operation cost of the i-th electrochemical energy storage system at the t-th moment; is the discharge power of the i-th electrochemical energy storage system at the t-th moment; is the charging power of the i-th electrochemical energy storage system at the t-th moment; The number of demand responses in the dispatch area; is the cost coefficient of the ith demand response at time t; is the i-th actual demand response at time t.

4. The partition joint peak scheduling method considering the day-ahead operation safety domain according to claim 3 is characterized in that: The overall constraints include: wind and solar safe operation range constraints, thermal power unit technical output upper and lower limits and group ramp rate constraints, energy storage system operation constraints, power balance constraints, line flow safety constraints and tie line constraints, system positive and negative spinning reserve constraints and demand response model constraints; The wind and solar power safe operation range constraints include: ； in, is the minimum output of the i-th wind turbine at the t-th moment; is the maximum output of the i-th wind turbine at time t; is the minimum output of the i-th photovoltaic unit at the t-th moment; is the maximum output of the i-th photovoltaic unit at the t moment; The technical output upper and lower limits of the thermal power unit and the group climbing rate constraints include: ； ； in, is the minimum output of the i-th thermal power unit at the t-th moment; is the maximum output of the i-th thermal power unit at time t; is the actual output of the i-th thermal power unit at time t+1; is the ramp power limit of the i-th thermal power unit; is the minimum output of the i-th thermal power unit at the t+1th moment; is the start / stop state variable of the i-th thermal power unit at the t+1th moment; is the maximum output of the i-th thermal power unit at time t+1; is the ramp-down power limit of the i-th thermal power unit; The energy storage system operation constraints include: ； in, is the power lower limit of the i-th electrochemical energy storage system at the t-th moment; is the power upper limit of the i-th electrochemical energy storage system at the t-th moment; is the charge state of the i-th electrochemical energy storage system at the time t+1; is the charge state of the i-th electrochemical energy storage system at the t-th moment; is the charging efficiency of the i-th electrochemical energy storage system at the t-th moment; is the discharge efficiency of the i-th electrochemical energy storage system at the t-th moment; is the tth scheduling period; is the rated energy of the i-th electrochemical energy storage system; is the low value of the state of charge of the i-th electrochemical energy storage system at the t-th moment; is the high value of the state of charge of the i-th electrochemical energy storage system at the t-th moment; The power balance constraint includes: ； in, is the i-th local load at time t; is the outgoing load of the jth tie line at time t; The number of tie lines for external load transmission; The line flow safety constraints and tie line constraints include: ； ； in, is the upper power limit of the lth line; is the actual power of line No. l at time t; is the upper limit of the power of the jth tie line; is the actual power of the jth tie line; The positive and negative spinning reserve constraints of the system include: ； ； in, is the total positive spinning reserve capacity at time t; is the total negative spinning reserve capacity at time t; is the upper reserve capacity factor of the wind turbine; is the upper reserve capacity factor of the PV unit; is the upper reserve capacity factor of the local load; is the lower reserve capacity factor of the wind turbine; is the lower reserve capacity factor of the PV unit; is the lower reserve capacity factor of the local load; The demand response model constraints include: ； in, is the lower limit of the ith demand response; is the peak value of the i-th local load before the day; is the maximum coefficient of the i-th demand response.

5. The partition joint peak scheduling method considering the day-ahead operation safety domain according to claim 4 is characterized in that: The currently operating security domain includes: ； in, It is the set of operating points of all units in the partition; It is the key decision vector corresponding to all the unit operation points in the partition within the dispatch cycle; is the discrete variable of the partitioned wind-solar-thermal-storage system; is the vector of the remaining variables of the system; is the power flow equation constraint; The unit combination constraint set includes: the wind and solar power safe operation range constraint, the upper and lower limits of the thermal power unit technical output and the group climbing rate constraint, the energy storage system operation constraint and the demand response model constraint; is the security constraint set; The expression of the power flow equality constraint is: ； in, is the number of thermal power units in the xth zone; is the number of wind turbines in the xth zone; is the number of PV panels in the xth zone; is the number of electrochemical energy storage systems in the x-th partition; is the number of local loads in the xth partition; is the number of tie lines in the xth zone; The expression of the safety constraint set is: ； in, is the actual power of the lth line in the xth partition at the tth moment; is the power upper limit of the lth line in the xth partition; is the actual power of the jth tie line in the xth partition at the tth moment; is the upper power limit of the j-th tie line in the x-th partition at the t-th moment; is the positive spinning reserve capacity of the xth partition at time t; is the negative spinning reserve capacity of the xth partition at time t.

6. The partition joint peak scheduling method considering the day-ahead operation safety domain according to claim 5 is characterized in that: The calculation formula of the upward adjustment capability includes: ； The calculation formula of the downward adjustment capability includes: ； in, is the upward adjustment capacity of the x-th partition at time t; is the downward adjustment capacity of the x-th partition at the t-th moment.

7. The partition joint peak scheduling method considering the day-ahead operation safety domain according to claim 6 is characterized in that: The peak demand allocation model includes: ； ； ； in, is the load demand of the newly added j-th tie line; is the number of partitions; Support power for the peak demand of the xth partition at time t; The power upper limit of the No. 1 tie line in the xth zone; is the power transfer distribution factor between the xth partition and the tie line No. 1; is the power transfer distribution factor between the lth tie line and the jth tie line; is the power flow of the lth line at the tth moment under the day-ahead scheduling plan; is the power upper limit of the jth tie line at the tth moment; is the lower limit of the regulation capacity of the xth partition at the tth moment; is a binary state variable characterizing the regulation capacity mode of the x-th partition at time t; is the upper limit of the regulation capacity of the x-th partition at the t-th moment; Set a threshold for the number of mode switches for the xth partition; is the target value to characterize the equilibrium peak; is the downward adjustment capacity corresponding to the maximum upward adjustment capacity of the x-th partition at time t; is the maximum upward adjustment capacity of the xth partition at time t; is the maximum downward adjustment capacity of the x-th partition at time t; is the upward adjustment capacity corresponding to the maximum downward adjustment capacity of the x-th partition at the t-th moment.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the partitioned joint peak scheduling method considering the day-ahead running safety domain as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the partition joint peak scheduling method considering the day-ahead running safety domain as described in any one of claims 1-7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the partition joint peak scheduling method considering the day-ahead running safety domain as described in any one of claims 1-7 is implemented.

Images