Cooperative optimization method and device for active power and frequency in black start of power system

Abstract

The invention provides a collaborative optimization method and device for active power and frequency in black start of a power system, and relates to the technical field of operation control. The method comprises the following steps: at the beginning of a first preset period, obtaining an active power output plan of each partition according to a recovery scheme, load prediction data and output prediction data of each partition; wherein the black start recovery process comprises a plurality of periods, and the first preset period is any period in the black start recovery process; updating the load prediction data and the output prediction data of each partition according to the active output plan of each partition every first preset time interval in a first preset period, and obtaining an active output reference value of each unit in each partition; and obtaining an active power compensation instruction according to the active output reference value of each unit and the real-time frequency deviation of each partition. According to the invention, accurate control can be realized, and the problem of uncontrollable frequency deviation caused by the fact that power change in the black start process cannot be tracked in real time is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of operation control, in particular to a method and device for coordinated optimization of active power and frequency in black start of a power system. BACKGROUND

[0002] With large-scale access of new energy and deepening of power grid interconnection and intercommunication, the difficulty of active power balance and frequency control in the process of black start and recovery of a power system is significantly improved, and the traditional recovery strategy has been difficult to meet the stability and rapidity requirements under complex operation conditions. Especially in the black start recovery stage after the whole system power failure.

[0003] A modern power system adopts a highly interconnected wide area network structure. Once a total outage accident occurs, the system parts need to be recovered in turn by the self-starting power generation units. The traditional active-frequency control strategy often relies on static start-up plans and simplified models, and is difficult to adapt to the modern power system recovery scenarios with high proportion of new energy, low inertia and strong disturbance characteristics. In the recovery process, the inertia level of the system is significantly reduced, and the strong fluctuation characteristics of the load and renewable power further amplify the frequency deviation, resulting in insufficient adjustment bandwidth of the conventional Automatic Generation Control (AGC) unit and even degradation to manual control, and the frequency control effect is severely limited. SUMMARY

[0004] Embodiments of the present application provide a method and device for coordinated optimization of active power and frequency in black start of a power system, to solve the problem that the traditional control strategy is difficult to track the power change in real time in the black start process, causing uncontrollable frequency deviation.

[0005] In a first aspect, embodiments of the present application provide a method for coordinated optimization of active power and frequency in black start of a power system, comprising: At the beginning of a first preset period, according to the recovery scheme, load prediction data and output prediction data of each partition, the active output plan of each partition is obtained; wherein the black start recovery process includes multiple periods, and the first preset period is any one period in the black start recovery process; In the first preset period, every first preset time interval, according to the active output plan of each partition, the load prediction data and output prediction data of each partition are updated, and the active output reference value of each unit in each partition is obtained; According to the active output reference value of each unit and the real-time frequency deviation of each partition, an active power compensation instruction is obtained.

[0006] In a possible implementation, the output prediction data includes new energy equipment output prediction data and traditional equipment output prediction data; the active power output plan of each partition is obtained according to the recovery scheme of each partition, the load prediction data and the output prediction data, and includes: At least one subsystem is determined according to the tie lines between the partitions; For any one subsystem, the minimum network loss of each partition included in the subsystem in the black start recovery process is taken as a first objective function according to the recovery scheme of each partition included in the subsystem; The first objective function of each subsystem is solved based on the first constraint condition of each subsystem, the first objective function, and the load prediction data and the output prediction data of the partitions included in each subsystem, to obtain the active power output plan of each partition in each time period and the start-stop state of each unit.

[0007] In a possible implementation, the first constraint condition of each subsystem includes the first sub-constraint condition of each partition included in the subsystem and the constraint condition of the tie line of each partition; The first objective function of each subsystem is solved based on the first constraint condition of each subsystem, the first objective function, and the load prediction data and the output prediction data of the partitions included in each subsystem, to obtain the active power output plan of each partition in each time period and the start-stop state of each unit, including: The first objective function of each subsystem is solved according to the first sub-constraint condition of each partition included in each subsystem, the constraint condition of the tie line of each partition, the first objective function, and the load prediction data and the output prediction data of the partitions included in each subsystem, to obtain the active power output plan of each partition in each time period and the start-stop state of each unit.

[0008] In a possible implementation, the first objective function of each subsystem is solved according to the first sub-constraint condition of each partition included in each subsystem, the constraint condition of the tie line of each partition, the first objective function, and the load prediction data and the output prediction data of the partitions included in each subsystem, to obtain the active power output plan of each partition in each time period and the start-stop state of each unit, including: The first preset period is divided into a plurality of optimization time periods; For any one subsystem, the following steps are performed: In each optimization time period, the first objective function of each subsystem is solved based on the first sub-constraint condition of each partition included in the subsystem, the constraint condition of the tie line of each partition, the first objective function, and the load prediction data and the output prediction data of the partitions included in the subsystem; According to the solution result, the active power output plan of each subarea and the start-stop state of each unit in each optimization period are obtained.

[0009] In a possible implementation, in the first preset period, the load prediction data and the output prediction data of each subarea are updated according to the active power output plan of each subarea and the active power reference value of each unit in each subarea is obtained every first preset time interval. In the first preset period, the load prediction data and the output prediction data of each subarea are updated according to the active power output plan of each subarea and the start-stop state of each unit. The active power reference value of each unit in each subarea is obtained based on the updated load prediction data and the output prediction data.

[0010] In a possible implementation, the active power reference value of each unit in each subarea is obtained based on the updated load prediction data and the output prediction data, and includes: The minimum power loss of each subarea is taken as a second objective function; The second objective function of each subarea is solved according to the updated load prediction data, the output prediction data of each period, and the second constraint condition of each subarea, and the active power reference value of each unit in each subarea is obtained.

[0011] In a possible implementation, the second objective function of each subarea is solved according to the updated load prediction data, the output prediction data of each period, and the second constraint condition of each subarea, and the active power reference value of each unit in each subarea is obtained, including: The second objective function of each subarea is solved according to the updated load prediction data, the output prediction data of each period, and the second constraint condition of each subarea, and the active power reference value corresponding to each time period in the first preset time interval is obtained. The active power reference value corresponding to the first time period in the first preset time interval is taken as the active power reference value of each unit in each subarea.

[0012] In a possible implementation, the active power compensation instruction is obtained according to the active power reference value of each unit and the real-time frequency deviation of each subarea, and includes: In the first preset period, the control deviation of each subarea is determined according to the real-time frequency deviation of each subarea every second preset time interval. The power control amount of each subarea is determined according to the control deviation of each subarea. The active power compensation instruction of each unit is obtained according to the power control amount of each subarea and the active power reference value. The second preset time interval is less than the first preset time interval.

[0013] In a second aspect, an embodiment of the present application provides a device for coordinated optimization of active power and frequency in a black start of a power system, comprising: a top layer module configured to, at the beginning of a first preset period, obtain active power output plans of each zone according to a recovery scheme of each zone, load prediction data and output prediction data; wherein the black start recovery process comprises a plurality of periods, and the first preset period is any one of the periods in the black start recovery process; a middle layer module configured to, in the first preset period, update the load prediction data and the output prediction data of each zone and obtain active power output reference values of each unit in each zone according to the active power output plans of each zone at every first preset time interval.

[0014] a bottom layer module configured to obtain active power compensation instructions according to the active power output reference values of each unit and real-time frequency deviations of each zone.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the method in the first aspect or any possible implementation manner of the first aspect when executing the computer program.

[0016] In the embodiment of the present application, the active power output plans of each zone are predicted in advance on the basis of the recovery scheme of each zone, the load prediction data and the output prediction data, the passive response in the black start recovery stage in the traditional method is changed into active response, the hysteresis in the process of regulation and control is avoided, the power system can be planned in advance, and risks in the recovery process can be coped with. Then, the load prediction data and the output prediction data of each zone are updated according to the active power output plans, the accuracy of the load prediction data and the output prediction data is ensured. Next, the active power output reference values of each unit in each zone are obtained according to the updated load prediction data and output prediction data, the power deviation between the active power output plans and actual operation can be corrected. Finally, the active power compensation instructions are determined according to the power change by analyzing the active power output reference values and the real-time frequency deviations of each zone, so as to compensate the active power in real time in the process of power adjustment of each unit, not only the accurate control can be realized, but also the problem that the frequency deviation is uncontrollable due to the failure to track the power change in the black start process in real time can be avoided. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is an implementation flowchart of the method for coordinated optimization of active power and frequency in a black start of a power system provided by the embodiment of the present application; Figure 2 is a control framework diagram of the method for coordinated optimization of active power and frequency in a black start of a power system provided by the embodiment of the present application; Figures 3a-3c is an effectiveness verification comparison chart of the active power and frequency collaborative optimization method in the black start of the power system provided by the embodiment of the application; Figure 4 is a frequency deviation comparison chart of each partition in different periods of the active power and frequency collaborative optimization method in the black start of the power system provided by the embodiment of the application; Figure 5 is a structural schematic diagram of the active power and frequency collaborative optimization device in the black start of the power system provided by the embodiment of the application; Figure 6 is a schematic diagram of the electronic device provided by the embodiment of the application. DETAILED DESCRIPTION

[0018] The embodiments of the application will be described in detail below with reference to the accompanying drawings.

[0019] Figure 1 is an implementation flowchart of the active power and frequency collaborative optimization method in the black start of the power system provided by the embodiment of the application. As shown in the figure, Figure 1 the method can include: Step 110: At the beginning of the first preset period, the active power output plan of each partition is obtained according to the recovery scheme, the load prediction data and the output prediction data of each partition; wherein the black start recovery process includes multiple periods, and the first preset period is any one period in the black start recovery process.

[0020] The black start recovery process often lasts for more than several hours, in order to realize accurate control, the black start recovery process is divided into multiple periods in the embodiment, and the corresponding control strategy is determined in each period, so as to improve the control accuracy.

[0021] The traditional method adopts centralized control, and all the devices in the target power system are uniformly regulated and controlled during control. This regulation and control method lacks flexibility in the early stage of black start recovery. Therefore, in order to overcome the defects of the traditional centralized control method, the scheme adopted in the embodiment acquires the recovery scheme, the load prediction data and the output prediction data of each partition according to the partition of the power system and taking each region as a unit when acquiring data.

[0022] Since the traditional method is difficult to adapt to the modern power system recovery scene with high proportion of new energy, low inertia and strong disturbance characteristics, the method provided in the embodiment considers the problems brought about by the new energy scene when planning control. Therefore, the output prediction data includes not only the output prediction data of traditional devices such as thermal power units and storage batteries, but also the output prediction data of new energy devices such as photovoltaic and wind power devices.

[0023] The control logic in the embodiment can be upper-layer control, which can make a prediction based on corresponding data in any one period of the black start recovery process, obtain long-period active power output plan in the period in advance, provide advanced data as a basis for subsequent optimization process, change passive response in the black start recovery stage in the traditional method into active response, avoid hysteresis in regulation, and enable the power system to plan in advance and cope with risks in the recovery process.

[0024] Step 120: In the first preset period, every first preset time interval, the load prediction data and the output prediction data of each partition are updated according to the active power output plan of each partition, and the active power reference value of each unit in each partition is obtained.

[0025] Since the traditional optimization control method often uses a single time scale, that is, the traditional control strategy often focuses on a certain time scale, such as minute-level scheduling or second-level automatic adjustment, and cannot coordinate the dynamic contradiction between long-period planning and short-period execution.

[0026] To solve this problem, the control logic in the embodiment can be middle-layer control, which dynamically adjusts the long-period active power output plan in the first preset period, dynamically corrects the load prediction data and the output prediction data of each partition for updating, timely responds to the prediction error of new energy equipment output and load, and dynamically updates the active power reference value of each unit in each partition on this basis to provide accurate reference for adjusting the frequency deviation.

[0027] Step 130: Obtain the active power compensation instruction according to the active power reference value of each unit and the real-time frequency deviation of each partition.

[0028] To realize long-period active power output plan determination, middle-period active power reference value adjustment, and short-period active power regulation, the control logic in the embodiment can be lower-layer control, which considers the real-time frequency deviation of each partition caused by the random fluctuation of new energy resources and load error when regulating, obtains the real-time power deviation according to the determined real-time power deviation, and combines the active power reference value to compensate for unstable new energy active power in the normal recovery of the target power system, avoid frequency deviation, and effectively improve the frequency stability in the recovery process.

[0029] In summary, the embodiment of the present application adopts a multi-time scale collaborative optimization control strategy, combines long-period active power output planning, medium-period active power reference value adjustment, and short-period active power regulation, and solves the problem that the traditional control strategy cannot coordinate the dynamic contradiction between long-period planning and short-period execution. In this process, by predicting the active power output plan of each subarea in advance, the hysteresis in regulation and control is avoided, the power system can be planned in advance to cope with risks in the recovery process, the active power output reference value of each unit in each subarea is dynamically updated to correct the power deviation between the active power output plan and the actual operation, and finally, by analyzing the active power reference value and the real-time frequency deviation of each subarea, the power change is quickly tracked, the active power compensation instruction is determined according to the power change, the active power is compensated in real time during the power adjustment of each unit, accurate control is realized, and the problem of uncontrollable frequency deviation due to the inability to track the power change in the black start process in real time is avoided.

[0030] In an optional embodiment, the output prediction data includes new energy device output prediction data and traditional device output prediction data; and the active power output plan of each subarea obtained in step 110 according to the recovery scheme of each subarea, the load prediction data and the output prediction data can include: At least one subsystem is determined according to the tie lines between the subareas.

[0031] For any one subsystem, the minimum network loss of each subarea included in the subsystem in the black start recovery process is taken as the first objective function according to the recovery scheme of each subarea included in the subsystem.

[0032] The first objective function of each subsystem is solved based on the first constraint condition, the first objective function of each subsystem, and the load prediction data and the output prediction data of each subarea included in each subsystem, to obtain the active power output plan of each subarea in each time period and the start-stop state of each unit.

[0033] If a traditional decentralized control method is used, the control ability of the obtained control strategy is weak, it is difficult to overall plan the global resource allocation, and the overall frequency control efficiency is reduced. Therefore, the centralized-decentralized collaborative control method is adopted in the embodiment.

[0034] When the target power system is recovered, a plurality of subareas are determined according to the subarea scheme of the target power system, and for each subarea, the active power output plan of each subarea is predicted based on the recovery scheme of the subarea and in combination with the load prediction data and the output prediction data of each subarea.

[0035] Specifically, according to the restoration scheme of each partition, a tie line between each partition in each time period is determined, wherein the tie line is used to represent the associated partitions in the restoration process. According to the time period corresponding to the tie line, each partition is arranged into at least one subsystem at the beginning of the time period.

[0036] For example, if partition a is arranged into partition b through the tie line in time period s, then from time period s, partition a and partition b are unified as a new subsystem for active power optimization. As the restoration process proceeds, after each partition is stably arranged through the tie line, the target power system enters a full-network unified restoration phase; when all key devices are restored to a normal state, the control strategy is switched from the restoration phase to a regular operation strategy, realizing full-network active power coordinated control. That is, in the restoration process, the number of subsystems gradually increases, and after increasing to the maximum number, the number gradually decreases, and finally each partition becomes a large system, which is the target power system.

[0037] For any one subsystem, the first constraint condition and the first objective function of the subsystem are unified as an upper-level optimization model of the subsystem, and are solved to concentrate on the coordination relationship between the partitions, and finally the obtained active power output plan is the active power output plan of each partition, and each partition is controlled individually to realize decentralized control.

[0038] The objective function of any one subsystem is the first objective function of each partition included in the subsystem, wherein the first objective function takes the minimum network loss of each partition as the target to reduce the network loss. The first constraint condition in the subsystem includes the first sub-constraint condition of each partition included in the subsystem and the constraint condition of the tie line of each partition; wherein the first sub-constraint condition of each partition includes a first node power balance constraint, a first reactive power constraint, a first active power constraint, a first branch power flow constraint, a first node voltage constraint, and a first branch transmission power constraint; the constraint condition of the tie line of each partition includes a power flow power balance constraint and a transmission capacity constraint.

[0039] For each subsystem, the first objective function and the first constraint condition in each subsystem are comprehensively solved. Since the time periods of the combination of each subsystem are different, the upper-level optimization model is essentially a multi-time period optimization problem, and the optimization problems corresponding to each time period are independent of each other. Therefore, in the solving process, the first preset period can be divided into multiple optimization time periods according to the combination time periods of each subsystem, and according to the multiple optimization time periods, the upper-level optimization model can be decomposed into upper-level optimization models of multiple optimization time periods, and each time period corresponding to the upper-level optimization model can be abstracted as a nonlinear programming problem, and each time period is solved respectively. In the solving process, a solver can be used for solving to obtain the active power output plan of each partition in each time period and the start-stop state of each unit.

[0040] In the embodiment, the first objective function can be expressed as:

[0041] wherein, is the objective function of the upper optimization model in the subarea ; is the number of subareas, which is determined by the system subarea scheme; is the number of total time periods of the restoration scheme, which is determined by the system active power restoration plan; is the set of node numbers in the subarea , which is determined by the system subarea plan; , , represents the active power plan value of the node to the node and the node to the node in the time period ; when there is a branch between the node and the node , i.e. the network loss of the branch in the time period .

[0042] The first node power balance constraint can be expressed as:

[0043] wherein, , , , and respectively represent the set of numbers of the conventional unit , the continuous reactive power compensation device , the discrete reactive power compensation device , the new energy station and the load connected to the node ; wherein the continuous reactive power compensation device includes static reactive power compensator, static reactive power generator and the like, and the discrete reactive power compensation device includes grouped capacitor / inductor group and the like; represents the set of numbers of the node connected to the node in the subarea , which is determined by the system subarea plan; and respectively represent the conventional unit and the new energy station that have been restored in the time period The active power output values ​​are determined by the system's active power recovery plan; Indicates the time period Restored load The active power demand values ​​are determined by the system's active power recovery plan; , , and These represent the time periods. Regular units that have been restored Continuous reactive power compensation device Discrete reactive power compensation device New energy power stations and load The planned value of reactive power output; Indicates the time period Restored load Planned reactive power demand; Represents a node and nodes Power between.

[0044] The first reactive power constraint can be expressed as:

[0045] In the formula, , , , and Representing regions All conventional units Continuous reactive power compensation device Discrete reactive power compensation device New energy power stations and load A set of numbers, which is determined by the system partitioning plan; , , Indicates conventional units Continuous reactive power compensation device and discrete reactive power compensation device During the period The status variable indicating whether recovery has been achieved is determined by the system's active power recovery plan. and They are conventional units Maximum and minimum limits for reactive power output; and These are continuous reactive power compensation devices. Maximum and minimum limits for reactive power output; Discrete reactive power compensation device During the period the planned value of the number of groups of capacitors / reactors to be put into operation; the maximum value of the number of groups of capacitors / reactors to be put into operation for the discrete reactive power compensation device ; the reactive power compensation value of a single group of capacitors / reactors to be put into operation for the discrete reactive power compensation device ; , the power factor angle of the new energy station and the load respectively.

[0046] The first active constraint can be expressed as:

[0047] In the formula, and are the maximum and minimum limits of the active power output of the conventional unit ; is the available power prediction value of the renewable power station r at the time period t; and are the upper and lower limits of the unit ramp rate.

[0048] The first branch power flow constraint can be expressed as:

[0049] In the formula, , are the real part and imaginary part of the element in the first row and the first column of the admittance matrix of the regional subsystem node respectively; represents the active power planned value of the node flowing to the node at the time period ; , are the voltage amplitudes of the node and the node at the time period ; is the voltage phase angle difference between the node and the node at the time period . The first node voltage constraint can be expressed as:

[0050]

[0051] In the formula, is the state variable of whether the node is powered on at the time period , which is determined by the system active power restoration plan;​​ and are the nodes maximum and minimum limits of the voltage.

[0052] The first branch transmission power constraint can be expressed as:

[0053] wherein, is the set of branch numbers in the area which is determined by the system partition plan; is the maximum limit of the apparent power of the branch is the state variable of whether the branch is energized or not, and the values are determined by the system active power recovery plan; is the maximum limit of the apparent power of the branch .

[0054] In an optional embodiment, in step 120, within the first preset period, the load prediction data and the output prediction data of each partition are updated according to the active power output plan of each partition and the active power output reference value of each unit in each partition is obtained every first preset time interval, which can include: Step 121: Within the first preset period, the load prediction data and the output prediction data of each partition are updated every first preset time interval according to the active power output plan of each partition and the start-stop state of each unit.

[0055] Step 122: The active power output reference value of each unit in each partition is obtained based on the updated load prediction data and output prediction data.

[0056] In order to realize fine control of each unit in each partition, the first preset period is divided into a plurality of first preset time intervals in the embodiment, and within the first preset period, the load prediction data and the output prediction data of each partition are updated every first preset time interval on the basis of the active power output plan of each partition, and the active power output reference value of each unit in the corresponding partition is obtained based on the updated load prediction data and output prediction data of each partition.

[0057] In this way, the accuracy of control can be improved, and a reference basis can be provided for subsequent regulation and control.

[0058] In an optional embodiment, in step 122, the active power output reference value of each unit in each partition is obtained based on the updated load prediction data and output prediction data, which can include: The minimum power loss of each partition is taken as the second objective function.

[0059] Solving the second objective function of each sub-zone according to the updated load prediction data, output prediction data, and second constraint condition of each sub-zone, to obtain the active power output reference value of each unit in each sub-zone.

[0060] In the embodiment, to obtain the active power output reference value of each unit in each sub-zone, it is required to minimize the power loss of the sub-zone while meeting the power balance and constraint condition in the sub-zone. Therefore, in the optimization process, the power loss of each sub-zone can be minimized as the second objective function, and the second node power balance constraint, the second reactive power constraint, the second active power constraint, the reserve capacity constraint, the second branch power flow constraint, the second node voltage constraint, and the second branch transmission power constraint can be taken as the second constraint condition of each sub-zone.

[0061] The second objective function and the second constraint condition can be taken as the middle-layer optimization model, the corresponding prediction data is updated every first time interval, and the middle-layer optimization model is solved once every first time interval. In the solving process, a solver is used to solve the second objective function of each sub-zone according to the updated load prediction data, output prediction data, and second constraint condition of each sub-zone, to obtain the active power output reference value corresponding to each time period in the first preset time interval, that is, the active power output reference value corresponding to each prediction window. Then, to ensure accuracy, only the active power output reference value at the beginning of the prediction window is taken as the active power output reference value of each unit in each sub-zone, that is, only the active power output reference value corresponding to the first time period in each time period in the first preset time interval is taken as the active power output reference value of each unit in each sub-zone. In this way, rolling closed-loop control is realized.

[0062] In an optional embodiment, the second objective function is:

[0063] In the formula, is the second objective function of the middle-layer optimization model in the area . is the number of time periods of the short-term optimization of the sub-zone, that is, the number of first time intervals in the first preset period; is the time length of each first time interval; is the number of restored nodes in the sub-zone; represents the load weight coefficient connected with the node i , represents the active power demand of the load connected with the node in the first i time interval. k

[0064] The second node power balance constraint is: ​

[0065] The second reactive power constraint is:

[0066] The second active power constraint is:

[0067] In the formula, is the active power output value of the unit g in the time period k, which is determined by the global active power lead planning layer. g is the active power output value of the unit g in the time period k, which is determined by the dynamic adjustment layer. k The active power output plan value of the unit g in the time period k is determined by the global active power lead planning layer. and represent the upper and lower limits of the decision of the dynamic adjustment layer which can be adjusted on the basis of the lead planning layer.

[0068] The reserve capacity constraint is:

[0069] In the formula, is the maximum output of the unit g, is the minimum reserve capacity requirement of the time period k.

[0070] The second branch power flow constraint is:

[0071] The second node voltage constraint is:

[0072] The second branch transmission power constraint is:

[0073] In an optional embodiment, the step 130 of obtaining the active power compensation instruction according to the active power reference value of each unit and the real-time frequency deviation of each sub-zone can include: In the first preset period, every second preset time interval, the control deviation of each sub-zone is determined according to the real-time frequency deviation of each sub-zone.

[0074] The power control amount of each sub-zone is determined according to the control deviation of each sub-zone.

[0075] The active power compensation instruction of each unit is obtained according to the power control amount of each sub-zone and the active power reference value.

[0076] The second preset time interval is less than the first preset time interval.

[0077] As the final link of the implementation control, the active power reference value of each unit is taken as the basis, and the real-time power deviation of each unit is corrected in combination with the real-time frequency deviation of each unit.

[0078] Specifically, compared with the upper control and the middle control, the situation of each area and even the whole target power system needs to be considered to determine the corresponding data. As the lower control, the embodiment only corrects according to the implementation frequency deviation of the subarea itself according to the situation of the subarea itself during control.

[0079] Wherein, the control deviation of each subarea is determined according to the real-time frequency deviation of each subarea by the following formula:

[0080] In the formula, is the control deviation is the area frequency deviation coefficient (MW / Hz), is the real-time power deviation.

[0081] The obtained is sent to the local PI controller corresponding to each subarea to obtain the power control amount of each subarea, so that the lower layer makes an active compensation according to the control deviation in each second time interval to quickly suppress the frequency deviation of each subarea. Wherein, the frequency control amount of real-time control is calculated by the following formula:

[0082] In the formula, is the power to be adjusted, and are the proportional coefficient and integral coefficient of PI control respectively, is the real-time sampling value of ACE, is the integral value of ACE.

[0083] Then, the embodiment does not directly change the active power reference value of each unit, but superimposes the obtained power control amount of each subarea and the active power reference value to obtain the active power compensation instruction of each unit, so as to distribute the active power compensation instruction of each unit to the AGC unit for execution.

[0084] To sum up, the embodiment of the present application divides the black start recovery process of the target power system into three levels, namely, an upper layer, a middle layer and a lower layer, and the time scales of each layer are different, so as to coordinate the dynamic contradiction between long-period planning and short-period execution. In actual planning, the active power output plan of each subarea is predicted in advance through the upper layer. The middle layer updates the load prediction data and the output prediction data of each subarea according to the active power output plan of each subarea of the upper layer, and obtains the active power output reference value of each unit in each subarea according to the updated data, thereby providing a reference for the lower layer. When the lower layer performs compensation control, the power to be compensated is determined based on the local real-time frequency deviation of the lower layer, so as to realize the suppression of the second-level frequency deviation and the active tracking. In addition, in the control process, the output of the new energy is reasonably predicted considering the volatility of the new energy, so as to effectively cope with the frequency deviation caused by the volatility of the new energy.

[0085] Figure 2 is the control framework diagram of the active power and frequency collaborative optimization method in the black start of the power system provided by the embodiment of the present application; the following will be described in combination with Figure 2 .

[0086] The control framework provided by the embodiment of the present application is a three-layer hierarchical structure, including an upper layer, a middle layer and a lower layer. The upper layer is a “global active power advance planning layer”, which divides the black start recovery process into multiple periods, and for any one period, the generator unit output and the load recovery rhythm are planned globally according to the real-time prediction data (including load prediction and output prediction) and the recovery scheme of each subarea, so as to determine the active power output plan of each subarea, wherein thirty minutes can be taken as a period.

[0087] The middle layer is a regional active power dynamic adjustment layer, and in this period, each subarea is recalibrated every first preset time interval, for example, every five minutes, according to the active power output plan obtained by the upper layer, the load prediction and the output prediction are corrected, and based on the corrected load prediction and output prediction, the active power output is adjusted to obtain the active power output reference value.

[0088] The lower layer is a local frequency automatic control layer, and every second preset time interval, for example, 8 seconds, the specific power control amount is determined according to the real-time frequency deviation of each subarea, and each subarea is controlled according to the power control amount and the active power output reference value obtained by the middle layer.

[0089] By constructing the centralized-distributed active power / frequency control system of “advance planning-rolling rescheduling-real-time rapid regulation”, the information interaction and collaborative optimization of multiple time scales are realized, and the safety, economy and recovery success rate can be considered in each stage of system recovery.

[0090] The effectiveness of the method proposed in the embodiment of the present application is verified through an optional embodiment.

[0091] The proposed method for coordinated optimization of active power and frequency in black start of a power system is verified according to actual data of a provincial power grid in a certain region. The power grid contains 280 substations, 34 thermal power plants, 1 pumped storage power station, 37 centralized wind power plants and 174 centralized photovoltaic power plants, 419 nodes of 220 kV and above, and 559 branches.

[0092] (1) System partitioning and system restoration scheme The system is divided into three regional subsystems, and each region is configured with one black start unit. The restoration plan uses 30 minutes as a single time period step, and the total restoration time is 6 hours. According to the generated system partitioning restoration scheme, the three regions respectively cover 111, 147 and 161 nodes of 220 kV and above. Table 1 lists the number of planned restoration nodes and the planned restoration load percentage of each region in each restoration period.

[0093] Table 1 Overview of system partitioning and restoration plan

[0094] As can be seen from Table 1, the three regions complete the black start restoration of their respective regional subsystems in the 10th, 10th and 11th periods, respectively. The regional 1 subsystem and the regional 2 subsystem complete system parallel operation in the 11th period, and the regional 3 subsystem and the merged system of regional 1 and regional 2 complete system parallel operation in the 12th period. The large system completes black start restoration in the 12th period.

[0095] (2) Effectiveness verification of active power / frequency control method To verify the effectiveness of the proposed three-layer active power-frequency control framework, the present application designs a comparative test: in the control scheme, a "two-layer" scheme containing only a global advance planning layer and a local second-level PI control layer is used as a comparison, i.e. the upper layer and the lower layer. The comparison focuses on evaluating the role of the 5min rolling rescheduling in the middle layer in absorbing new energy output fluctuations and load prediction errors, thereby reducing power outage losses.

[0096] Figures 3a-3c is a comparison chart for effectiveness verification of the method for coordinated optimization of active power and frequency in black start of a power system provided by the embodiment of the present application; wherein, Figure 3a is a comparison chart for effectiveness verification of regional 1, Figure 3b is a comparison chart for effectiveness verification of regional 2, Figure 3cThe comparison chart for the effectiveness verification of region 3; in each chart, the blue line represents the proportion of the recovered load obtained by using the two-layer framework, and the red line represents the proportion of the recovered load obtained by using the three-layer framework.

[0097] It can be seen from the comparison of the three charts that the three-layer scheme significantly improves the load recovery percentage at all stages, thereby improving the economy and frequency safety of the system recovery process.

[0098] Figure 4 is the comparison chart of the frequency deviation of each subarea at different stages of the active power and frequency collaborative optimization method provided by the embodiment of the power system black start; from Figure 4 It can be seen that the frequency deviation at each stage is limited within a small range, and according to the black start technical specification, the maximum frequency deviation of different regions at each stage during the black start process does not exceed ±0.5Hz, which meets the technical requirements of black start, so the control strategy obtained by using the method provided by the embodiment of the application in the black start recovery stage completely meets the technical requirements.

[0099] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the application.

[0100] The following is a device embodiment of the application, and for details not described in detail, reference can be made to the corresponding method embodiments described above.

[0101] Figure 5 The structure schematic diagram of the active power and frequency collaborative optimization device in the power system black start provided by the embodiment of the application is shown, only the parts related to the embodiment of the application are shown for convenience of description, and the details are as follows: As Figure 5 shown, the active power and frequency collaborative optimization device 5 in the power system black start comprises: The upper layer module 51 is configured to obtain the active power output plan of each subarea according to the recovery scheme of each subarea, the load prediction data and the output prediction data at the beginning of the first preset period; wherein, the black start recovery process comprises a plurality of periods, and the first preset period is any one of the periods in the black start recovery process; The middle layer module 52 is configured to update the load prediction data and the output prediction data of each subarea according to the active power output plan of each subarea every first preset time interval within the first preset period, and obtain the active power output reference value of each unit in each subarea.

[0102] The lower layer module 53 is configured to obtain the active power compensation instruction according to the active power output reference value of each unit and the real-time frequency deviation of each subarea.

[0103] In a possible implementation, the output prediction data includes new energy equipment output prediction data and traditional equipment output prediction data; and the upper module 51 is specifically configured to: determine at least one subsystem according to the tie lines between the subareas; for any one of the subsystems, take the minimum network loss of the subareas included in the subsystem in the black start recovery process as a first objective function according to the recovery schemes of the subareas included in the subsystem; based on the first constraint condition of each subsystem, the first objective function, and the load prediction data and the output prediction data of the subareas included in each subsystem, solve the first objective function of each subsystem to obtain the active power output plan of each subarea in each time period and the start-stop state of each unit.

[0104] In a possible implementation, the first constraint condition of each subsystem includes the first sub-constraint condition of the subareas included in the subsystem and the constraint condition of the tie lines of the subareas; The upper module 51 is specifically configured to: based on the first sub-constraint condition of the subareas included in each subsystem, the constraint condition of the tie lines of the subareas, the first objective function, and the load prediction data and the output prediction data of the subareas included in each subsystem, solve the first objective function of each subsystem to obtain the active power output plan of each subarea in each time period and the start-stop state of each unit.

[0105] In a possible implementation, the upper module 51 is specifically configured to: divide the first preset period into multiple optimization time periods; for any one of the subsystems, the following steps are performed: in each optimization time period, based on the first sub-constraint condition of the subareas included in the subsystem, the constraint condition of the tie lines of the subareas, the first objective function, and the load prediction data and the output prediction data of the subareas included in the subsystem, solve the first objective function of each subsystem; according to the solving result, obtain the active power output plan of each subarea in each optimization time period and the start-stop state of each unit.

[0106] In a possible implementation, the middle module 52 is specifically configured to: in the first preset period, update the load prediction data and the output prediction data of each subarea according to the active power output plan of each subarea and the start-stop state of each unit every first preset time interval; based on the updated load prediction data and the output prediction data, obtain the active power output reference value of each unit in each subarea.

[0107] In a possible implementation, the middle-layer module 52 is specifically configured to: take the minimum power loss of each sub-zone as a second objective function; solving the second objective function of each sub-zone according to the updated load prediction data, the output prediction data of each time period, and the second constraint condition of each sub-zone, to obtain the active power output reference value of each unit in each sub-zone.

[0108] In a possible implementation, the middle-layer module 52 is specifically configured to: solving the second objective function of each sub-zone according to the updated load prediction data, the output prediction data of each time period, and the second constraint condition of each sub-zone, to obtain the active power output reference value corresponding to each time period in the first preset time interval; taking the active power output reference value corresponding to the first time period in the first preset time interval as the active power output reference value of each unit in each sub-zone.

[0109] In a possible implementation, the lower-layer module 53 is specifically configured to: determining the control deviation of each sub-zone according to the real-time frequency deviation of each sub-zone in the first preset period and every second preset time interval; determining the power control amount of each sub-zone according to the control deviation of each sub-zone; obtaining the active power compensation instruction of each unit according to the power control amount and the active power output reference value of each sub-zone; wherein the second preset time interval is less than the first preset time interval.

[0110] Figure 6 is a schematic diagram of an electronic device provided by an embodiment of the present application. As shown in the figure, the electronic device 6 of this embodiment includes a processor 60 and a memory 61. The memory 61 stores a computer program 62. The processor 60 implements the steps in each of the method embodiments described above when executing the computer program 62. Alternatively, the processor 60 implements the functions of each module / unit in each of the device embodiments described above when executing the computer program 62. Figure 6 By way of example, the computer program 62 can be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 62 in the electronic device 6.

[0111] The electronic device 6 can include, but is not limited to, the processor 60 and the memory 61. Those skilled in the art can understand that,

[0112] Figure 6 ​The electronic device 6 is merely an example and does not constitute a limitation on the electronic device 6, and can include more or fewer components than illustrated, or combine certain components, or different components, for example, the electronic device 6 can also include an input / output device, a network access device, a bus, etc.

[0113] For the convenience and brevity of description, only the above-mentioned division of each functional module / unit is exemplified, and in actual application, the above-mentioned functions can be completed by different functional modules / units according to needs. The above-mentioned modules / units can be realized in the form of hardware, software or a combination of hardware and software.

[0114] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in a certain embodiment can be referred to the related description of other embodiments. If there is no special description and logical conflict, the terms and / or descriptions of different embodiments are consistent and can be mutually referred to, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0115] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for coordinated optimization of active power and frequency in black start of a power system, characterized in that, The method comprises the following steps: At the beginning of a first preset period, active power output plans of each sub-zone are obtained according to the recovery scheme of each sub-zone, the load prediction data and the output prediction data; wherein the black start recovery process comprises a plurality of periods, and the first preset period is any one of the periods in the black start recovery process; In the first preset period, the load prediction data and the output prediction data of each sub-zone are updated according to the active power output plans of each sub-zone, and active power output reference values of each unit in each sub-zone are obtained at every first preset time interval; Active power compensation instructions are obtained according to the active power output reference values of each unit and real-time frequency deviations of each sub-zone.

2. The method for coordinated optimization of active power and frequency in black start of power system according to claim 1, characterized in that, The output prediction data comprises new energy equipment output prediction data and traditional equipment output prediction data; the active power output plans of each sub-zone are obtained according to the recovery scheme of each sub-zone, the load prediction data and the output prediction data, which comprises the following steps: At least one subsystem is determined according to the tie lines between the sub-zones; For any one of the subsystems, the minimum network loss of each sub-zone included in the subsystem in the black start recovery process is taken as a first objective function according to the recovery scheme of each sub-zone included in the subsystem; The first objective function of each subsystem is solved based on the first constraint condition of each subsystem, the first objective function, and the load prediction data and the output prediction data of the sub-zones included in each subsystem, to obtain the active power output plans of each sub-zone at each time period and the start-stop state of each unit.

3. The method for coordinated optimization of active power and frequency in black start of power system according to claim 2, characterized in that, The first constraint condition of each subsystem comprises the first sub-constraint condition of each sub-zone included in the subsystem and the constraint condition of the tie lines of each sub-zone; The first objective function of each subsystem is solved based on the first constraint condition of each subsystem, the first objective function, and the load prediction data and the output prediction data of the sub-zones included in each subsystem, to obtain the active power output plans of each sub-zone at each time period and the start-stop state of each unit. The first objective function of each subsystem is solved based on the first constraint condition of each subsystem, the first objective function, and the load prediction data and the output prediction data of the sub-zones included in each subsystem, to obtain the active power output plans of each sub-zone at each time period and the start-stop state of each unit.

4. The method for coordinated optimization of active power and frequency in black start of power system according to claim 3, characterized in that, The first preset period is divided into a plurality of optimization time periods; For any one of the subsystems, the following steps are performed: In each optimization time period, the first objective function of each subsystem is solved based on the first sub-constraint condition of each sub-zone included in the subsystem, the constraint condition of the tie lines of each sub-zone, the first objective function, and the load prediction data and the output prediction data of the sub-zones included in the subsystem; In each optimization time period, the first objective function of each subsystem is solved based on the first sub-constraint condition of each sub-zone included in the subsystem, the constraint condition of the tie lines of each sub-zone, the first objective function, and the load prediction data and the output prediction data of the sub-zones included in the subsystem; According to the solving result, the active power output plan of each partition and the start-stop state of each unit in each optimization period are obtained.

5. The method for coordinated optimization of active power and frequency in black start of power system according to claim 2, characterized in that, The active power output reference value of each unit in each partition is obtained based on the updated load prediction data and output prediction data. The active power output reference value of each unit in each partition is obtained based on the updated load prediction data and output prediction data. The active power output reference value of each unit in each partition is obtained based on the updated load prediction data and output prediction data.

6. The method for coordinated optimization of active power and frequency in black start of power system according to claim 5, characterized in that, The active power output reference value of each unit in each partition is obtained based on the updated load prediction data and output prediction data. The minimum power loss of each partition is taken as a second objective function. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition.

7. The method for coordinated optimization of active power and frequency in black start of power system according to claim 6, characterized in that, The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition.

8. The method for coordinated optimization of active power and frequency in black start of power system according to claim 1, characterized in that, The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition.

9. A device for coordinated optimization of active power and frequency in black start of a power system, characterized in that, The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in each partition is obtained by solving the second objective function of each partition according to the updated load prediction data, output prediction data and second constraint condition of each partition. The active power output reference value of each unit in 10. An electronic device, comprising: A computer program product comprising a computer readable medium having stored thereon the computer program of claim 9, wherein said computer program is configured such that, when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is carried out. A computer program product comprising a computer readable medium having stored thereon the computer program of claim 9, wherein said computer program is configured such that, when the computer program is executed by a processor, the

Images