Adaptive multi-target cooperative scheduling method for power distribution area based on flexible interconnection device

Abstract

The invention relates to a flexible interconnection device-based self-adaptive multi-target cooperative scheduling method for a power distribution area, and belongs to the technical field of electric power. The method takes minimization of voltage fluctuation and flexible load fluctuation of an AC / DC power distribution system as a target, an optimization target function is constructed, and meanwhile, SOP power constraint, system power flow constraint, system operation safety constraint, energy storage constraint and flexible load constraint are taken as constraint conditions; dividing the operation mode of the transformer area, and then respectively constructing dynamic weight coefficients in a steady-state operation mode, a voltage early warning mode and a secondary voltage recovery mode; and solving the optimization objective function, and scheduling according to a solving result. According to the method, multi-target collaborative optimization is realized, and the economic operation and source load fluctuation impact coping capability of each transformer area in the power distribution transformer area are improved; by detecting voltage fluctuation in real time, the adjusting function of a controllable device can be fully played, the system power flow can be rapidly and accurately adjusted, and economical and reliable operation of the system is achieved.

Description

Technical Field

[0001] This invention belongs to the field of power technology, specifically relating to an adaptive multi-objective cooperative scheduling method for distribution substations based on flexible interconnection devices. Background Technology

[0002] Under the "dual-carbon" strategy, new energy power generation will gradually replace traditional fossil fuels, and the construction of a new power grid dominated by new energy is in full swing. A high proportion of new energy and a high proportion of power electronic devices are important directions for power system development and core characteristics of power system development. This will bring about significant changes in the corresponding power system theories, system architecture, operation methods, and technologies.

[0003] Limited by factors such as short-circuit capacity and electromagnetic ring networks, traditional power distribution systems are mostly designed as closed loops but operate in open loops. Their power flow changes with network structure parameters and load variations, and can only be regulated by switching control. This "passive" physical architecture and operating mode results in weak power supply reliability and power control capabilities, easily leading to line capacity overload and node voltage exceeding limits, severely restricting the connection of various power sources, loads, and energy storage devices.

[0004] With the rapid development of power electronics technology, intelligent devices capable of interconnecting distribution networks and flexibly adjusting power flow are gradually entering different levels of distribution networks. The use of this new type of equipment, through flexible interconnection between distribution substations, improves the flexibility of the power grid and ensures the safe and reliable operation of the distribution network, providing an effective means to solve the aforementioned problems. Existing research collectively refers to this intelligent equipment with flexible access capabilities as flexible interconnection devices. Based on existing research results, using Flexible Interconnection Devices (FIDs) to achieve flexible interconnection of distribution networks can effectively improve the operational quality of the distribution network, representing a promising new distribution network structure. Building on this, current research proposes the concept of Flexible Interconnected Distribution Networks (FDNs), integrating technologies such as Soft Open Points (SOPs) and Flexible Multi-state Switches (FMSs) into the distribution network to replace traditional mechanical switching methods, improving the control flexibility and reliability of the distribution network, and providing the basic conditions and core equipment for flexible load routing in the distribution network.

[0005] Currently, for AC / DC distribution networks that achieve flexible interconnection based on power electronic devices, there are mainly two control modes: centralized and distributed. Centralized control places higher demands on strong communication capabilities, but the large number and wide distribution of various power supply units in the system, coupled with limitations imposed by communication speed, leads to a decrease in the system's power supply reliability. Distributed control is based on local information, and therefore cannot achieve optimal energy scheduling from a global perspective. Although existing research has established a central controller to regulate the output between various units, it is affected by the system's communication rate because it is a centralized control system. Furthermore, most control modes adopt master-slave control under normal conditions, while distributed control is used when the grid experiences a fault, and this switching requires real-time communication. Some research has proposed a control method based on voltage margin, in which, when a constant DC voltage converter station fails, the backup converter station can switch from constant power control to droop control and participate in grid power balancing. However, this research only focuses on the coordination of individual links in the system and does not consider the overall power scheduling of the entire system. Current hierarchical control methods for multi-terminal DC distribution networks include a first-level step-down mechanism, a second-level voltage recovery mechanism, and a third-level tie-line current mechanism, but they do not consider the coordination between upper and lower levels. The aforementioned research status indicates that some exploration has been conducted on the control problem of flexible interconnected AC / DC distribution networks, but most of these efforts focus on the coordinated operation of components within the system, without considering the system's global optimal operation and coordinated control.

[0006] Therefore, overcoming the shortcomings of existing technologies is an urgent problem to be solved in the field of power technology. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide an adaptive multi-objective cooperative scheduling method for distribution radio areas based on Flexible Interconnect Device (FID).

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] The adaptive multi-objective cooperative scheduling method for distribution radio areas based on flexible interconnection devices includes the following steps:

[0010] Step (1) aims to minimize voltage fluctuations and flexible load fluctuations in the AC / DC power distribution system. An optimization objective function is constructed, with the power constraints of the SOP, power flow constraints of the system, system operation safety constraints, energy storage constraints, and flexible load constraints as constraints.

[0011] The objective function is:

[0012] minF=β1(t)f′1+β2(t)f′2+β3(t)f′3 (1)

[0013] In the formula, f′1, f′2, and f′3 represent the normalized objective function values ​​of system operating cost, voltage deviation, and flexible load fluctuation, respectively; β1(t) is the weighting coefficient of system operating cost; β2(t) is the weighting coefficient of voltage deviation; and β3(t) is the weighting coefficient of flexible load fluctuation.

[0014] Step (2) divides the transformer area operation mode into the following three typical states: steady-state operation mode, voltage warning mode and secondary voltage recovery mode;

[0015] Dynamic weighting coefficients are constructed for steady-state operation mode, voltage warning mode, and secondary voltage recovery mode, respectively;

[0016] Step (3) involves real-time detection of voltage fluctuations on the high-voltage side of the main transformer in the distribution area, calculation of dynamic weighting coefficients under the corresponding distribution area operation mode, solving the optimization objective function, and then controlling the distribution area according to the solution results.

[0017] Furthermore, preferably, in step (1), the system operating cost f1 is composed of the electricity purchase cost C. pur Costs of wind and solar power curtailment (C) DG Power loss cost C loss And the cost of energy storage charging and discharging C ESS composition;

[0018] f1 = C pur +C DG +C loss +C ESS (2)

[0019]

[0020] In the formula, P(t) represents the electricity price at time t; P grid (t) represents the active power interaction value between the distribution area and the upper-level power grid at time t; c DG This indicates the penalty price for wind and solar power curtailment; P′ wt and P′ pv P represents the active power generated by the wind turbine and the photovoltaic generator, respectively. m,wt (t) and P m,pv (t) represents the actual active power utilized by the m-th wind turbine and the m-th photovoltaic generator at time t, respectively, where m = 1, 2, 3, ..., n; ij (t) and R ij (t) represent the current and resistance on branch ij within the distribution area at time t, respectively, and φ1 represents the set of all branches within the distribution area; K ess This represents the unit charge / discharge cost coefficient for energy storage; r a y and y represent the depreciation rate and service life of energy storage, respectively; P h,ch and Ph,dis These represent the charging and discharging power of the h-th energy storage device, respectively.

[0021] Furthermore, preferably, in step (1), the specific calculation method for the voltage deviation f2 is as follows:

[0022]

[0023] In the formula, U i (t) represents the high-voltage side voltage of the i-th distribution transformer at time t, n k This indicates the total number of stations.

[0024] Furthermore, preferably, in step (1), the specific calculation method for the flexible load fluctuation f3 is as follows:

[0025]

[0026] In the formula, ΔP load P represents the change in flexible loads in a distribution network. d in (t), P u tr (t) and P d tr (t) are all decision variables for user-side scheduling, representing the planned power of the load that can be cut off, the increased load, and the decreased load that can be transferred at time t.

[0027] Furthermore, preferably, in step (1),

[0028] The power constraints of SOP are as follows:

[0029]

[0030] In the formula, and These represent the active power injected by SOP into nodes i and j during time period t, respectively. and These represent the reactive power injected by SOP into nodes i and j during time period t, respectively. and These represent the active power losses of the voltage source converters (VSCs) connected to node i and node j during time period t, respectively. and These represent the loss coefficients of SOP and VSC corresponding to node i and node j, respectively. and These represent the minimum and maximum reactive power injected by the VSC connected to node i, respectively. and These represent the minimum and maximum reactive power injected by the VSC connected to node j, respectively. and These represent the capacity limits of the VSC connected to node i and the VSC connected to node j, respectively.

[0031] The specific power flow constraints of the system are:

[0032]

[0033] In the formula, P ij (t) and Q ij (t) represent the active and reactive power flowing through branch ij within the distribution area at time t, respectively; Ω i Let ψ represent the set of transformer substations starting from substation i. i R represents the set of transformer substations ending at substation i; li and X li These represent the resistance and reactance on branch line li of the transformer substation, respectively; I li (t) represents the current value flowing through branch li in the transformer area at time t; U l (t) and U i (t) represent the voltage amplitudes at nodes l and i in the distribution area at time t, respectively; P i inj and Q i inj These represent the active and reactive power injected into the transformer area, respectively. The calculation expression is:

[0034]

[0035] In the formula, P i,DG (t) and Q i,DG (t) represent the active and reactive power injected by DG into transformer area i at time t, respectively; P i,SOP (t) and Q i,SOP (t) represent the active and reactive power injected by SOP at time t, respectively; and P represents the charging and discharging power of the i-th distribution station area at time t, respectively; i,Load_a (t) and Q i,Load (t) represents the active and reactive power required by the load at time t, respectively.

[0036] Furthermore, preferably, in step (1), the system operation security constraints are specifically as follows:

[0037]

[0038] In the formula, I ij (t) represent the current in branch ij of transformer area at time t; I ij,max U represents the maximum allowable current in branch ij; i(t) represents the effective voltage value of node i at time t; U i,min U represents the minimum allowable voltage at node i. i,max This represents the maximum allowable voltage at node i.

[0039] Furthermore, preferably, in step (1), the energy storage constraint is specifically as follows:

[0040] E h,min ≤E h (t)≤E h,max (19)

[0041] In the formula, E h,max and E h,min E represents the upper and lower limits of the energy storage capacity of the h-th energy storage device, respectively; h The SOC value of the h-th energy storage device is expressed as follows:

[0042]

[0043] In the formula, E h (t) represents the SOC value of the energy stored in the h-th energy storage device at time t, E h (t-1) represents the SOC value of the h-th energy storage device at the previous moment, i.e., t-1; P h ch (t) represents the energy storage charging power of the h-th energy storage device at time t; P h dis (t) represents the energy storage discharge power of the h-th energy storage device at time t; η ch and η dis Let represent the charge and discharge conversion efficiencies of the h-th energy storage device, respectively; Δt represents the scheduling time interval; to ensure the absolute continuity of the stored energy in time, the following also needs to be satisfied:

[0044] E h (0)=E h (24) (21)

[0045] In the formula, E h (0) represents the initial SOC value of the h-th energy storage device at time 0; E h (24) represents the SOC value of the energy stored by the h-th energy storage device at the end of the cycle, i.e., at time 24;

[0046] Energy storage charging and discharging constraints are:

[0047]

[0048] In the formula, and These represent the maximum energy storage charging and discharging power of the h-th energy storage device, respectively. and These represent the charging and discharging states of the energy stored in the h-th energy storage device.

[0049] Furthermore, preferably, in step (1), the flexible load constraint specifically refers to:

[0050] When flexible loads participate in system scheduling, the power demand of the loads is:

[0051]

[0052] In the formula, P i,Load_a (t) represents the active power required by the load at time t; P i,Load (t) represents the active power required by the load before the transfer at time t; P d in (t), P u tr (t) and P d tr (t) are decision variables for user-side scheduling, representing the planned power of the load that can be cut off, the increased load, and the decreased load that can be transferred at time t, respectively.

[0053] The constraints that transferable loads must satisfy are:

[0054]

[0055] In the formula, P u tr (t) represents the planned power of the transferable load increased by the user side at time t using the method; P d tr (t) represents the planned power of the transferable load reduced on the user side at time t; Indicates the upper limit of transferable load;

[0056] The constraints that the load can be removed must satisfy are:

[0057]

[0058] In the formula, P in (t) represents the planned load that can be cut off on the user side at time t; and These represent the upper and lower limits of the removable load, respectively.

[0059] Furthermore, preferably, in step (2):

[0060] Stable operation mode: When the voltage deviation is kept within ±5% of the rated voltage, the system is in steady-state operation mode, and the main goal is to optimize economic efficiency.

[0061] Voltage warning mode: When the voltage deviation is within ±5% to ±7% of the rated voltage, the system enters voltage warning mode, and voltage stability should be given priority at this time.

[0062] Secondary voltage recovery mode: When the voltage deviation exceeds ±7% of the rated voltage, the system enters the secondary voltage recovery mode, at which time the primary goal is to quickly restore voltage stability.

[0063] Furthermore, preferably, in step (2), the expression for the dynamic weighting coefficient under steady-state operation mode is:

[0064]

[0065] In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; ε1 and ε2 represent the linearization coefficients of the weighting coefficients, ε1 represents the basic weight, and ε2 represents the sensitivity coefficient to voltage deviation; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max This represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode;

[0066] The expression for the dynamic weighting coefficient in voltage warning mode is:

[0067]

[0068] In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max U represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode; N This indicates the rated voltage value of the main transformer in the distribution substation area;

[0069] The expression for the dynamic weighting coefficient in secondary voltage recovery mode:

[0070]

[0071] In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; ε1 and ε2 represent the linearization coefficients of the weighting coefficients, ε1 represents the basic weight, and ε2 represents the sensitivity coefficient to voltage deviation; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU maxThis represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode.

[0072] In step (1) of this invention, β1(t) is the weighting coefficient of system operating cost. In steady-state operation mode, voltage fluctuation is small. At this time, the optimization strategy prioritizes economy and is set to a higher value. β2(t) is the weighting coefficient of voltage deviation. In early warning mode, as voltage deviation increases, the main goal of the optimization strategy is to maintain the system voltage quality deviation back to steady-state operation. At this time, the coefficient is increased to prioritize voltage stability. β3(t) is the weighting coefficient of flexible load fluctuation. In order to highlight the impact of flexible load scheduling on voltage deviation, this invention ignores the flexible load pricing process and uses flexible load power fluctuation to represent user satisfaction. In secondary voltage recovery mode, since the voltage deviation has exceeded the safe operating range, the weighting coefficient is adjusted by a linear decreasing function based on the degree of voltage deviation to quickly restore voltage stability.

[0073] In step (1) of this invention, P grid (t) represents the active power interaction value between the distribution area and the upper-level power grid at time t, indicating the power absorbed from the main grid.

[0074] I ij (t) and R ij (t) represents the current and resistance of branch ij (branch number in the distribution substation, indicating the line from node i to node j) within the substation area at time t, respectively; φ1 represents the set of all branches within the distribution substation area; r a y and y represent the depreciation rate (0.08) and applicable life (15 years) of energy storage, respectively.

[0075] Voltage warning mode: When the voltage deviation is within ±5% to ±7% of the rated voltage (inclusive), the system enters voltage warning mode, and voltage stability should be given priority at this time;

[0076] In step (2) of this invention, ε1 and ε2 represent the linearization coefficients of the weighting coefficients. ε1 represents the basic weight, which reflects the inherent sensitivity of the system to voltage deviation; ε2 represents the sensitivity coefficient to voltage deviation, which emphasizes the voltage optimization target and can be tuned according to the voltage optimization requirements of the power distribution system. In this invention, ε1 and ε2 are preferably taken as 0.3 and 0.4. max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max This represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode.

[0077] In step (3) of this invention, the voltage fluctuation of the high-voltage side of the main transformer in the distribution area is detected in real time, and the corresponding dynamic weighting coefficient is calculated. By solving the optimization objective function, the optimal output of each controllable device and the flexible load control scheme can be obtained. Then, the flexible interconnection devices and flexible loads between distribution areas are scheduled according to the obtained control scheme, thereby realizing multi-objective collaborative optimization of the distribution network operation economy, voltage fluctuation and power supply reliability.

[0078] This invention analyzes the requirements of different operating modes and formulates an optimized scheduling framework adapted to the system's operating state, achieving multi-objective collaborative optimization. This improves the economic operation of each distribution area and its ability to cope with source-load fluctuations. Furthermore, the adaptive weighting coefficient determination method proposed in this invention, through real-time voltage fluctuation detection, can fully utilize the regulatory role of controllable devices to quickly and accurately adjust the system power flow, achieving economical and reliable system operation. When the voltage deviation is small, this method can improve the economic efficiency of system operation; when the voltage deviation is large, it can reduce voltage fluctuations and improve the system's power supply reliability.

[0079] Compared with the prior art, the beneficial effects of this invention are as follows:

[0080] (1) Significantly improved multi-objective collaborative optimization capability. This invention achieves collaborative optimization of economy, voltage stability, and flexible load fluctuation through a dynamic weight coefficient adjustment mechanism. Simulation results show that, under steady-state operation mode, the system operating cost is reduced to 15,853 yuan (compared to 16,001 yuan for the fixed weight strategy); under voltage warning mode, the operating cost is further optimized to 75,186 yuan (compared to 76,326 yuan for the existing strategy);

[0081] (2) Fast adaptive response and excellent convergence performance. Through dynamic weighting coefficient design and real-time voltage monitoring, the response time of the system in different operating modes is significantly shortened. For example, it only takes 19.2 seconds to converge in steady-state mode and 10.94 seconds in voltage warning mode, which is better than the traditional centralized control strategy (usually more than 30 seconds);

[0082] (3) Improved efficiency of flexible load regulation and enhanced user satisfaction. In the secondary voltage recovery mode, this invention significantly improves power supply reliability and user experience by dynamically adjusting the participation of flexible loads. Attached Figure Description

[0083] Figure 1 A schematic diagram showing the division of the operating status of the transformer substations;

[0084] Figure 2 This is a diagram of the optimized scheduling framework proposed in this invention;

[0085] Figure 3The simulation analysis uses an improved IEEE 33-node distribution network structure diagram; (dashed lines indicate smart soft switch SOP connections; SOP1: connects transformer substations 8 and 29; SOP2: connects transformer substations 17 and 24; SOP3: connects transformer substations 20 and 31)

[0086] Figure 4 Output curve of distributed power source;

[0087] Figure 5 The charging and discharging power of the energy storage device; ESS1 represents energy storage device 1, ESS2 represents energy storage device 2, and ESS3 represents energy storage device 3;

[0088] Figure 6 The output power of different ports of the smart soft switch (SOP) is shown below; (a) is the output power of the two VSC ports of SOP1 (VSC1 represents the connection area 8; VSC2 represents the connection area 29); (b) is the output power of the two ports of SOP2 (VSC1 represents the connection area 17; VSC2 represents the connection area 24); and (c) is the output power of the two VSC ports of SOP3 (VSC1 represents the connection area 20; VSC2 represents the connection area 31).

[0089] Figure 7 The voltage of the distribution area in 24 hours is given by different optimized scheduling strategies in Scheme 1; where (a) is the optimized scheduling strategy with fixed weight coefficients; and (b) is the optimized scheduling strategy proposed in this invention.

[0090] Figure 8 These are the weighting coefficients for different optimization objectives in Scheme 1;

[0091] Figure 9 These are the weighting coefficients for different optimization objectives in Scheme 2;

[0092] Figure 10 The 24-hour distribution area voltages under different optimized scheduling strategies in Scheme 2 are: (a) a fixed weight coefficient optimized scheduling strategy; (b) a single-mode adaptive optimized scheduling strategy; (c) an existing multi-mode operation scheduling strategy; and (d) the optimized scheduling strategy proposed in this invention.

[0093] Figure 11 These are the weighting coefficients for different optimization objectives in Scheme 3;

[0094] Figure 12 The 24-hour distribution area voltage is calculated using the fixed weight coefficient optimization scheduling strategy in Scheme 3; where (a) does not consider flexible loads; and (b) considers flexible loads.

[0095] Figure 13 The 24-hour distribution area voltage for the optimized scheduling strategy proposed in Scheme 3;

[0096] Figure 14 The flexible load fluctuation situation of different optimization scheduling strategies in Scheme 3 is shown; where (a) is the optimization scheduling strategy with fixed weight coefficient; and (b) is the optimization scheduling strategy proposed in this invention. Detailed Implementation

[0097] The present invention will now be described in further detail with reference to the embodiments.

[0098] Those skilled in the art will understand that the following embodiments are for illustrative purposes only and should not be construed as limiting the scope of the invention. Where specific techniques or conditions are not specified in the embodiments, they are performed in accordance with the techniques or conditions described in the literature in the field or according to the product instructions. Materials or equipment whose manufacturers are not specified are all conventional products that can be obtained by purchase.

[0099] To achieve optimal operation of AC / DC distribution networks based on flexible interconnection devices under normal conditions, it relies on the interaction and communication between the upper-level controller and the lower-level controlled objects, as well as among the controlled objects. This invention, based on measured data, load forecasts, and photovoltaic power generation predictions, considers multiple aspects such as economy, voltage stability, and system losses. It employs an optimal algorithm to solve for the energy dispatch commands of each power source, thereby realizing energy dispatch for each power generation unit.

[0100] Division of Operation Status of 1st Circuit

[0101] Since the high-voltage side of the transformer in the distribution area is at the starting point of the power transmission in the distribution area, its voltage fluctuation characteristics will be transmitted to each node along the internal transmission line of the distribution area, thereby affecting the risk of voltage exceeding the limit within the distribution area. Therefore, based on the voltage value of the high-voltage side of the main transformer in the low-voltage distribution area, this invention divides the operation mode of the distribution area into the following three typical states: steady-state operation mode, voltage warning mode, and secondary voltage recovery mode.

[0102] When the voltage deviation is within ±5% of the rated voltage, the system is in steady-state operation mode. In this mode, the transformer substation has excellent disturbance rejection capability and sufficient power regulation margin, and can withstand large power fluctuation impacts, thereby maintaining the voltage operation of the substation within a reasonable range.

[0103] When the voltage deviation exceeds ±7% of the rated voltage, the system enters the secondary voltage recovery mode. In this mode, the distribution area may experience relatively serious abnormal operating conditions such as instantaneous transformer overload and large power fluctuations caused by instantaneous output of the distribution generator.

[0104] When the voltage deviation is within ±5% to ±7% of the rated voltage (inclusive), although the system is within the allowable operating range, its operational reliability is significantly reduced, and even slight power fluctuations can trigger the risk of voltage exceeding limits. Therefore, it is necessary to establish a voltage deviation early warning mechanism and adopt preventive control strategies to keep voltage fluctuations within a safe range.

[0105] To improve the voltage quality of power distribution systems, this invention proposes a three-level operating state model of "steady state-early warning-recovery" suitable for low-voltage distribution substations, such as... Figure 1 As shown.

[0106] 2 Adaptive Optimization Scheduling Framework

[0107] The stochastic fluctuations in distributed generation (DG) output and load demand may cause the distribution network to operate in different modes at different times, and its optimization requirements will also change. Traditional optimization scheduling models are usually limited to power flow optimization under a single operating state, which is difficult to meet the operational requirements of new power systems. Therefore, it is necessary to construct an optimization scheduling framework applicable to multiple operating states based on the multimodal operating characteristics and optimization requirements of the distribution system, so as to achieve adaptive optimization scheduling under different operating modes.

[0108] When the low-voltage distribution substation group is in steady-state operation, the voltage fluctuations on the high-voltage side of each substation transformer remain within a small range, and the various controllable devices within the distribution substation have sufficient adjustability margins. At this time, the system possesses operating conditions aimed at optimal economic efficiency. Therefore, the main optimization scheduling strategy under this operating mode should be to minimize system operating costs.

[0109] When a low-voltage distribution transformer network enters voltage warning mode, the voltage on the high-voltage side of some transformers in the regional distribution network is approaching the critical threshold. At this time, the system exhibits high sensitivity to power fluctuations; even slight fluctuations in source and load power can cause the transformer voltage to exceed the steady-state operating range. Therefore, the optimized dispatching strategy in warning mode mainly focuses on reducing the range of transformer voltage fluctuations and improving the voltage quality of the transformer areas.

[0110] In the secondary voltage recovery mode, the voltage deviation of some distribution substation access nodes has exceeded the safe operating range. At this time, it is necessary to quickly adjust the voltage of the distribution substations through control strategies such as limiting DG output or reducing load, so that the operating status of the distribution substation group can be restored to the early warning mode or steady-state operation mode. Therefore, in this mode, economic optimization is no longer considered, but reducing voltage deviation is the primary optimization objective, while also taking into account the fluctuation of flexible loads on the user side.

[0111] Furthermore, to alleviate the problem of frequent power flow fluctuations in the distribution system caused by switching between different operating modes, this invention adopts a scheduling planning framework with a time scale of 1 hour and an optimization period of 24 hours. Based on the above analysis, the optimized scheduling framework adapted to multiple operating states proposed in this invention is as follows: Figure 2 As shown.

[0112] 3 Adaptive Multi-Objective Optimization Model

[0113] The adaptive multi-objective optimization model of this invention includes an optimization objective function and constraints, as detailed below:

[0114] (I) Optimization Objectives

[0115] The goal of the regional energy regulation center is to achieve optimal economic efficiency while ensuring system stability and improving the absorption of distributed energy resources. Therefore, the adaptive multi-objective optimization model of this invention aims to minimize voltage fluctuations and flexible load fluctuations in the AC / DC distribution system. The resulting objective function is:

[0116] minF=β1(t)f′1+β2(t)f′2+β3(t)f′3 (1)

[0117] In the formula, f′1, f′2, and f′3 represent the normalized objective function values ​​of system operating cost, voltage deviation, and flexible load fluctuation, respectively; β1(t) is the weighting coefficient of system operating cost; β2(t) is the weighting coefficient of voltage deviation; and β3(t) is the weighting coefficient of flexible load fluctuation.

[0118] 1) System operating cost f1

[0119] f1 is determined by the electricity purchase cost C. pur Costs of wind and solar power curtailment (C) DG Power loss cost C loss And the cost of energy storage charging and discharging C ESS composition.

[0120] f1 = C pur +C DG +C loss +C ESS (2)

[0121]

[0122] In the formula, P(t) represents the electricity price at time t; P grid (t) represents the active power interaction value between the distribution area and the upper-level power grid at time t; c DG This indicates the penalty price for wind and solar power curtailment; P′ wt and P′ pv P represents the active power generated by the wind turbine and the photovoltaic generator, respectively. m,wt (t) and P m,pv (t) represents the actual active power utilized by the m-th wind turbine and the m-th photovoltaic generator at time t, respectively, where m = 1, 2, 3, ..., n; ij (t) and R ij(t) represents the current and resistance of branch ij (branch number in the distribution substation, indicating the line from node i to node j) within the substation area at time t, respectively; φ1 represents the set of all branches within the distribution substation area; K ess This represents the unit charge / discharge cost coefficient for energy storage; r a y and y represent the depreciation rate (0.08) and applicable life (15 years) of energy storage, respectively; P h,ch and P h,dis These represent the charging and discharging power of the h-th energy storage device, respectively.

[0123] 2) Voltage deviation f2

[0124] To highlight the significant advantages of the proposed optimized scheduling strategy in reducing the risk of voltage exceedance, it is assumed that all transformer areas are in a state of three-phase voltage balance. Therefore, the optimization objective mainly considers voltage deviation.

[0125] To reduce the risk of voltage exceeding limits during transformer operation, this invention uses the absolute value of voltage deviation as the optimization target to minimize the deviation between the transformer node voltage and the rated voltage.

[0126]

[0127] In the formula, U i (t) represents the high-voltage side voltage of the i-th distribution transformer at time t, n k This indicates the total number of stations.

[0128] 3) Flexible load fluctuation f3

[0129] Transferable and controllable loads, as flexible load resources, participate in distribution network power flow optimization as independent entities after signing grid connection and dispatch agreements with the power grid operator. To highlight the impact of flexible load dispatch on voltage deviation, this invention ignores the flexible load pricing process and uses a simplified flexible load power model to optimize distribution network voltage on the user side.

[0130] When voltage fluctuations exceed the safe operating range, flexible loads are required to participate in the dispatching plan to ensure reliable system operation. Considering the impact of flexible load dispatching on user power comfort, this invention uses the power fluctuation of flexible loads to represent user satisfaction.

[0131] f3=ΔP load =|P u tr (t)-P d tr (t)-P d in (t)| (6)

[0132] In the formula, ΔP loadP represents the change in flexible loads in a distribution network. d in (t), P u tr (t) and P d tr (t) are all decision variables for user-side scheduling, representing the planned power of the load that can be cut off, the increased load, and the decreased load that can be transferred at time t using the method of this invention.

[0133] In order to accurately reflect the requirements of each optimization objective, the objective function needs to be normalized, that is:

[0134] f′ k =f k / f k,0 (7)

[0135] In the formula, f′ k f represents the normalized value of the optimization objective k, where k = 1, 2, 3; k f represents the value of the k-th optimization objective after optimization; k,0 This represents the value of the k-th optimization objective before optimization. "Before optimization" refers to the system's target value without the optimization scheduling strategy, reflecting the system's initial state or baseline operating condition. "After optimization" refers to the value after adopting the optimization scheduling strategy, adjusting resources such as energy storage, smart soft switching (SOP), and flexible loads, reflecting the actual effect of the optimization strategy.

[0136] (II) Constraints

[0137] To achieve coordinated optimization of DG, SOP, ESS, upper-level grid and flexible loads, flexible interconnected distribution substations need to satisfy the power constraint equation of SOP, system power flow constraints, system operation safety constraints, energy storage constraints and flexible load constraints during the dispatching process.

[0138] 1) SOP power constraint

[0139]

[0140] In the formula, and These represent the active power injected by SOP into nodes i and j during time period t, respectively. and These represent the reactive power injected by SOP into nodes i and j during time period t, respectively. and These represent the active power losses of the voltage source converters (VSCs) connected to node i and node j during time period t, respectively. and These represent the loss coefficients of SOP and VSC corresponding to node i and node j, respectively. and These represent the minimum and maximum reactive power injected by the VSC connected to node i, respectively. and These represent the minimum and maximum reactive power injected by the VSC connected to node j, respectively. and These represent the capacity limits of the VSC connected to node i and the VSC connected to node j, respectively.

[0141] 2) System power flow constraints

[0142]

[0143] In the formula, P ij (t) and Q ij (t) represent the active and reactive power flowing through branch ij within the distribution area at time t, respectively; Ω i Let ψ represent the set of transformer substations starting from substation i. i R represents the set of transformer substations ending at substation i; li and X li These represent the resistance and reactance on branch line li of the transformer substation, respectively; I li (t) represents the current value flowing through branch li in the transformer area at time t; U l (t) and U i (t) represent the voltage amplitudes at nodes l and i in the distribution area at time t, respectively; P i inj and These represent the active and reactive power injected into the transformer area, respectively. The calculation expression is:

[0144]

[0145] In the formula, P i,DG (t) and Q i,DG (t) represent the active and reactive power injected by DG into transformer area i at time t, respectively; P i,SOP (t) and Q i,SOP (t) represent the active and reactive power injected by SOP at time t, respectively; and P represents the charging and discharging power of the i-th distribution station area at time t, respectively; i,Load_a (t) and Q i,Load (t) represents the active and reactive power required by the load at time t, respectively.

[0146] As can be seen from the voltage deviation optimization objective (5), the voltage always tends to adjust towards the rated value during the optimization scheduling process. When the voltage deviation is greater than the reference value, the adaptive multi-objective optimization model of this invention can reduce the system voltage by reducing the DG utilization rate and coordinating the control of ESS and SOP to absorb excess power in the distribution area; when the voltage deviation is less than the reference value, the adaptive multi-objective optimization model can improve the system voltage by increasing the DG utilization rate and coordinating the control of ESS and SOP to compensate for the power in the distribution area. Therefore, the optimization scheduling model proposed in this invention can reduce voltage fluctuations and improve the voltage quality during system operation.

[0147] 3) System operation security constraints

[0148] To ensure the safe operation of the system, branch capacity constraints and node voltage constraints must be met. The specific expressions are:

[0149]

[0150] In the formula, I ij (t) represent the current in branch ij of transformer area at time t; I ij,max U represents the maximum allowable current in branch ij; i (t) represents the effective voltage value of node i at time t; U i,min U represents the minimum allowable voltage at node i. i,max This represents the maximum allowable voltage at node i.

[0151] 4) Energy storage constraints

[0152] Energy storage constraints are generally divided into capacity constraints and power constraints. The specific expression for capacity constraints is:

[0153] E h,min ≤E h (t)≤E h,max (19)

[0154] In the formula, E h,max and E h,min E represents the upper and lower limits of the energy storage capacity of the h-th energy storage device, respectively; h The SOC value of the h-th energy storage device is expressed as follows:

[0155]

[0156] In the formula, E h (t) represents the SOC value of the energy stored in the h-th energy storage device at time t, E h (t-1) represents the SOC value of the h-th energy storage device at the previous moment, i.e., t-1; P h ch(t) represents the energy storage charging power of the h-th energy storage device at time t; P h dis (t) represents the energy storage discharge power of the h-th energy storage device at time t; η ch and η dis Let represent the charge and discharge conversion efficiencies of the h-th energy storage device, respectively; Δt represents the scheduling time interval; to ensure the absolute continuity of the stored energy in time, the following also needs to be satisfied:

[0157] E h (0)=E h (24) (21)

[0158] In the formula, E h (0) represents the initial SOC value of the h-th energy storage device at time 0; E h (24) represents the SOC value of the energy stored by the h-th energy storage device at the end of the cycle, i.e., at time 24.

[0159] Energy storage charging and discharging constraints are:

[0160]

[0161] In the formula, and These represent the maximum energy storage charging and discharging power of the h-th energy storage device, respectively. and These represent the charging and discharging states of the energy stored in the h-th energy storage device.

[0162] 5) Flexible load constraints

[0163] When flexible loads participate in system scheduling, the power demand of the loads is:

[0164]

[0165] In the formula, P i,Load_a (t) represents the active power required by the load at time t; P i,Load (t) represents the active power required by the load before the transfer at time t; P d in (t), P u tr (t) and P d tr (t) are all decision variables for user-side scheduling, representing the planned power of the load that can be cut off, the increased load, and the decreased load that can be transferred, obtained by using the adaptive multi-objective cooperative scheduling method for distribution substations proposed in this invention at time t.

[0166] The constraints that transferable loads must satisfy are:

[0167]

[0168] In the formula, P u tr (t) represents the planned power of the transferable load increased by the user side using the method of this invention at time t; P d tr (t) represents the planned power of transferable load reduced by the user side using the method of the present invention at time t; Indicates the upper limit of transferable load;

[0169] The constraints that the load can be removed must satisfy are:

[0170]

[0171] In the formula, P in (t) indicates that the planned load power can be cut off by the user side using the method of the present invention at time t; and These represent the upper and lower limits of the removable load, respectively.

[0172] (III) Model Conversion

[0173] The low-voltage distribution substation optimization model constructed in this invention belongs to a typical non-convex nonlinear mixed integer programming problem, and its solution complexity increases exponentially with the problem size. To improve the model's solution efficiency, linearization is required. First, variables U'(t) and I'(t) are introduced to replace the squared term variable U in the original model. 2 (t) and I 2 (t) converts nonlinear constraints into linear constraints. This conversion method can shorten the model solution time and improve optimization efficiency while ensuring the accuracy of the optimization results.

[0174]

[0175] In the formula, P ij (t) and Q ij (t) represent the active and reactive power flowing through branch ij within the distribution area at time t, respectively; Ω i Let ψ represent the set of transformer substations starting from substation i. i R represents the set of transformer substations ending at substation i; li and X li These represent the resistance and reactance on branch line li of the transformer substation, respectively; I′ li (t) represents the square of the current flowing through branch li in the transformer area at time t; U′ l (t) and U′ l (t) represents the squared values ​​of the voltage amplitudes at nodes l and i in the distribution area at time t, respectively; P iinj and Q i inj These represent the active and reactive power injected into the transformer area, respectively.

[0176] Secondly, for the absolute value nonlinear term in the objective function, an auxiliary variable V is introduced. i (t), and construct the corresponding linearized constraint set to transform equation (5) into:

[0177]

[0178] Among them, V i (t) needs to satisfy:

[0179]

[0180] To address the constraints of non-convex nonlinearity that persist after model transformation, a second-order cone relaxation technique is employed. This method effectively reduces the complexity of the solution while ensuring that the feasible solution set of the original problem remains unchanged, and simultaneously guarantees that the optimization result has global optimality. Equation (13) can be transformed into:

[0181]

[0182] 4. Multi-objective weight determination method

[0183] The optimization model proposed in this invention pertains to a multi-objective optimization problem, and can achieve synergistic optimization among different optimization objectives through weighted summation. To reduce the influence of subjective factors on the optimization results, this invention proposes a dynamic adaptive weight coefficient determination method based on the hierarchical classification of system operating states.

[0184] (I) Steady-state operation mode

[0185] In steady-state operation mode, since the voltage deviation remains within a small range, the optimization strategy prioritizes improving system economy. However, to cope with potential voltage fluctuations, a linear weighting adjustment mechanism based on voltage deviation is adopted, so that the weighting coefficient of voltage deviation dynamically increases with the increase of voltage fluctuation. Considering the overall system operation optimization requirements, the weighting coefficient of the economic optimization objective should be relatively large, but will decrease with the increase of voltage deviation. Furthermore, since the voltage deviation is small, there is no need for load transfer or shelving in this mode. Therefore, user satisfaction is not considered in the optimization model. Based on the above analysis, by calculating the deviation between the node voltage on the 10kV side of the main transformer in the low-voltage distribution substation and the reference voltage value, the dynamic weighting coefficient expression in steady-state operation mode is constructed as follows:

[0186]

[0187] In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; ε1 and ε2 represent the linearization coefficients of the weighting coefficients, with ε1 representing the basic weight, reflecting the inherent sensitivity of the system to voltage deviation; ε2 represents the sensitivity coefficient to voltage deviation, emphasizing the voltage optimization objective. The parameters can be tuned according to the voltage optimization requirements of the power distribution system; in this invention, they are taken as 0.3 and 0.4; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max This represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode.

[0188] In steady-state operation mode, the system has sufficient economic adjustment margin, and the weight coefficients of different optimization objectives are determined according to equation (34). At this time, the system focuses on economic optimization, and when the maximum voltage deviation U of the transformer area is detected... max As β(t) increases, the voltage weighting coefficient β2(t) increases slowly, while the economic weighting coefficient β1(t) decreases synchronously, ensuring that the system operates economically while controlling voltage fluctuations within ±5%.

[0189] (II) Early Warning Operation Mode

[0190] As voltage deviation increases, when the system is in voltage warning mode, the primary optimization objective is to maintain the system's voltage quality and restore the voltage deviation to normal operating conditions. Therefore, improving voltage deviation should be the main optimization objective. To enhance the sensitivity of the weighting coefficients to voltage deviation, the expression for the dynamic weighting coefficients in the voltage warning mode constructed in this invention is as follows:

[0191]

[0192] In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max U represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode; N This indicates the rated voltage value of the main transformer in the distribution substation.

[0193] When U max(t) When the system exceeds the steady-state operating range, it enters the voltage warning mode, and the weighting coefficient is determined by equation (35). Since the optimization requirements of the system vary greatly under different operating modes, the linearization adjustment coefficient of the weighting coefficient is increased in equation (35) to improve the weighting adjustment rate, so that the weighting coefficient in the voltage warning mode follows the voltage deviation more obviously, and the value of the voltage weighting term is significantly increased compared with the normal operating mode.

[0194] (III) Secondary Voltage Recovery Mode

[0195] When the system enters the secondary voltage recovery mode, since the voltage deviation has exceeded the safe operating range, the system's economic efficiency is no longer considered. Instead, load-side coordinated dispatch is required to achieve voltage recovery. Simultaneously, to ensure power reliability on the user side, user satisfaction must be considered, and the weight of user satisfaction should gradually decrease as the voltage deviation increases. Therefore, an expression for the dynamic weighting coefficients in the secondary voltage recovery mode is obtained using a linear function based on the degree of voltage deviation:

[0196]

[0197] In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; ε1 and ε2 represent the linearization coefficients of the weighting coefficients, with ε1 representing the basic weight, reflecting the inherent sensitivity of the system to voltage deviation; ε2 represents the sensitivity coefficient to voltage deviation, emphasizing the voltage optimization objective. The parameters can be tuned according to the voltage optimization requirements of the power distribution system; in this invention, they are taken as 0.3 and 0.4; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max This represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode.

[0198] With U max As (t) further expands, the distribution area enters the secondary voltage recovery mode. At this time, the system can initiate an emergency control strategy according to equation (36) to introduce flexible loads to improve the voltage quality of the distribution network. At the same time, a dynamic adjustment model of the weight coefficients of voltage quality and user satisfaction is constructed to realize the trend of linear increase of the voltage quality weight coefficient as the voltage deviation increases, while the user satisfaction weight coefficient decreases synchronously. When the voltage deviation further expands to the point that the system cannot maintain normal operation, the system will automatically trigger the black start contingency plan (referring to the emergency recovery mechanism in which the power system automatically starts its own power supply and gradually restores power supply through existing preset intelligent criteria and processes when a serious fault or even a power outage is detected, so as to quickly restore the operation of the power grid), and rebuild the power supply network through a hierarchical and zonal recovery strategy to minimize the risk of large-scale power outages.

[0199] Based on the above analysis, according to the proposed multi-objective dynamic weight coefficient determination method, the system monitors the high-voltage side voltage value of the main transformer in different distribution substations in real time, adaptively selects the weight coefficient calculation model under the corresponding operating mode, realizes dynamic adjustment of the weight coefficient under different operating modes, meets the differentiated optimization needs of the distribution network under different operating modes, achieves coordinated optimization of system economy and voltage fluctuation, and effectively improves the adaptability and robustness of distribution network operation.

[0200] 5. Simulation Analysis

[0201] (I) Example Data

[0202] This invention uses an improved IEEE 33-node distribution network structure as a case study, such as... Figure 3 As shown in Table 1, node 33 is designated as the interconnection node of the upstream power grid (main grid), while the remaining nodes correspond to the high-voltage side access points of the 10kV / 400V transformers in different distribution areas. The system's base capacity is 1MV·A. Regarding DG configuration, distribution areas 17 and 23 are connected to wind turbine generators, while distribution areas 8, 20, and 32 are connected to photovoltaic power generation systems. Considering the net load of the distribution areas, the load balance of the interconnected distribution areas, voltage regulation requirements, and the relative distance between distribution areas, distribution areas (8,29), (17,24), and (20,31) are selected as SOP access points. The rated capacity of each port of the SOP is shown in Table 1. To enhance the system's voltage regulation capability, energy storage is configured in distribution areas 15, 20, and 32. Peak shaving and valley filling strategies are used to smooth voltage fluctuations and reduce the risk of system voltage exceeding limits. Under extreme operating conditions, when interconnected distribution areas simultaneously experience voltage exceeding limits, the system will coordinate the charging and discharging status of the energy storage with the output of DG to ensure that the voltage quickly recovers to the safe operating range.

[0203] Table 1

[0204]

[0205] To reduce the impact of distributed generation (DG) output uncertainty on dispatch results, this invention, based on the output models of photovoltaic (PV) systems and wind turbines, employs Latin hypercube sampling and scenario reduction to select the wind turbine and PV output curves with the highest scenario probabilities, and uses these curves as the predicted DG active power values. Figure 4 As shown.

[0206] To improve the accuracy and practical application value of the optimization model, a time-of-use pricing mechanism is introduced when calculating operating costs. By guiding load transfer and optimizing distributed generation (DG) output through price, the system operating costs are minimized. The electricity prices for each time period are shown in Table 2.

[0207] Table 2

[0208]

[0209]

[0210] (II) Adaptive Optimization Scheduling Results

[0211] Taking the optimized scheduling results under stable operation mode as an example, the output of the energy storage device obtained based on the optimized scheduling strategy proposed in this invention is as follows: Figure 5 As shown.

[0212] according to Figure 4 Power output diagrams of solar (PV) and wind (WT) at different times and Figure 5 Analysis of the energy storage device processing diagram reveals that during periods 4-6, wind turbines (WT) operate at low power output, and photovoltaic (PV) power generation is not outputting due to solar energy constraints. The power generation within the distribution network is insufficient to meet load demand. To reduce electricity purchase and power transmission losses, energy storage (ESS) releases a certain amount of electricity to meet the active power demand of the distribution area. During periods 7-12, the output of both PV and WT gradually increases. At this time, energy storage absorbs the surplus active power within the distribution area, compensating for the nighttime power output of the energy storage system and simultaneously enhancing distributed energy efficiency. The power generation and absorption level; during the period from 13 to 19, the output of photovoltaic (PV) power generation and wind turbine (WT) power generation gradually decreases, but their output power can still meet the power supply load demand in the distribution area. At this time, the electricity price is at a flat level. Considering economic factors, the power deficit is mainly obtained from the upper-level grid. During the period from 20 to 23, the output of photovoltaic (PV) power generation is 0. Energy storage (ESS) meets the power supply demand of the distribution area load by releasing electricity and reduces the voltage fluctuation of the distribution network. After 24 hours, the load demand decreases and the electricity price is low. Each distribution area mainly obtains power from the upper-level grid to meet the load operation demand.

[0213] The active power transmission curves over time for different SOPs are as follows: Figure 6As shown in the diagram, the converter (VSC1) ports of smart soft switches 1 (SOP1) and 3 (SOP3) are connected to distribution areas with high photovoltaic (PV) generator penetration. Therefore, during periods of sufficient sunlight, to improve DG absorption and reduce voltage fluctuations, the two distribution areas need to transmit a large amount of active power to the interconnected distribution areas via the smart soft switches (SOPs). The receiving distribution areas receive power to compensate for the power deficit during heavy loads, thus reducing the system's power loss costs and electricity purchase costs to some extent. The negative power at the VSC1 ports of SOP1 and SOP3 indicates the output power of the distribution areas via the SOPs, while the positive power at the converter (VSC2) ports within the smart soft switches indicates the input power of the distribution areas via the SOPs. The VSC1 port of SOP2 is connected to the distribution area with a high penetration rate of wind turbine units. During the peak output period of wind turbines, in order to avoid the risk of voltage exceeding the limit in the distribution area caused by reverse power flow, the redundant active power of distribution area 17 is transferred to the interconnected distribution area or other time periods through the coordinated control strategy of intelligent soft switch (SOP) and energy storage (ESS) to ensure that the system operates within a safe range.

[0214] The simulation results and analysis above show that the adaptive optimization scheduling strategy proposed in this invention can optimize system power loss and voltage level by controlling controllable units within the distribution area.

[0215] (III) Simulation Analysis of Different Operating Modes

[0216] To verify the effectiveness and superiority of the adaptive multi-objective optimization scheduling strategy proposed in this invention, a comparative experimental scheme was designed to analyze the performance differences of different scheduling schemes under three operating modes.

[0217] 1) Scheme 1: In steady-state operation mode, a fixed weight coefficient optimization scheduling strategy is adopted, that is, the weight ratio of system operating cost and voltage deviation is fixed at 1:1, and the system voltage deviation and operating cost are compared with those obtained by the optimization scheduling strategy proposed in this invention.

[0218] Figure 7 The operating voltage characteristics of two optimized scheduling strategies under steady-state operation mode over 24 hours are demonstrated. The results show that the optimized scheduling strategy proposed in this invention has significant advantages in improving voltage quality, and can reduce the maximum voltage fluctuation of the transformer access node to 0.0283.

[0219] Figure 8 The dynamic change process of the optimization target weight coefficient in the optimized scheduling strategy proposed in this invention is further presented. Since the system is in a stable operating mode, flexible loads do not need to participate in scheduling in order to ensure power supply reliability. Figure 8The simulation primarily reflects the coordinated adjustment characteristics of economic weight and voltage deviation weight. When the system experiences a significant voltage deviation within 20 hours, the voltage deviation weight coefficient increases instantaneously, while the economic weight decreases simultaneously. This dynamic adjustment mechanism effectively suppresses further voltage deterioration. Simulation results show that the proposed optimized scheduling strategy can respond promptly to the system's operating status and dynamically adjust the optimization target weights according to actual needs, thereby improving the power supply reliability of the system in steady-state operation mode.

[0220] According to the optimization results, the operating costs of the fixed-weight optimized scheduling strategy and the optimized scheduling strategy proposed in this invention within 24 hours are 16001 yuan and 15853 yuan, respectively. Figure 7 Voltage fluctuation characteristics of each distribution area and Figure 8 As can be seen from the dynamic change process of the weighting coefficients, when the voltage deviation in the distribution area is small, the optimized scheduling strategy proposed in this invention can fully utilize the power regulation margin of the flexible interconnected distribution area and reduce the system operating cost. When the voltage fluctuation in the distribution area intensifies, the optimized scheduling strategy proposed in this invention can significantly enhance the safe operation capability of the system by dynamically adjusting the weighting coefficients of voltage deviation and economic optimization objectives. Therefore, under steady-state operation mode, the proposed adaptive optimized scheduling strategy can achieve synergistic optimization of voltage quality and operational economy.

[0221] Furthermore, when the system is always running in steady-state operation mode, the proposed adaptive optimization scheduling strategy converges after 19.2 seconds after 9 iterations, with a final convergence interval of 0.0097%, which not only meets the time constraints of real-time scheduling of the distribution network, but also meets the requirements of the required convergence accuracy.

[0222] 2) Option 2: When the system enters a voltage warning mode, the following strategies are compared with the optimization scheduling strategy proposed in this invention: fixed weight coefficient optimization scheduling strategy (i.e., the weight ratio of system operating cost and voltage deviation is fixed at 1:1), single-mode adaptive optimization scheduling strategy (i.e., dynamic optimization method for a single operating mode, without involving coordination between multiple modes), and existing multi-mode operation scheduling strategy (i.e., monitoring voltage values ​​at different times to determine optimization targets at different times, but without considering the coordinated optimization of economy and voltage quality).

[0223] Under the fixed weight coefficient optimization scheduling strategy, when the load fluctuates significantly, the voltage in the distribution area will deviate significantly, causing the system to switch from steady-state operation mode to voltage warning mode. Figure 9 This demonstrates the dynamic change of the optimization target weights when the system is in voltage warning mode. At 21:00, the maximum voltage fluctuation of the system exceeds ±5% of the rated voltage. At this time, the weight coefficient of the voltage deviation is determined by the voltage warning mode, and the voltage deviation weight increases rapidly. However, the weight coefficient determination method of the steady-state operation mode is still used in other time periods.

[0224] Figure 10 The voltage fluctuations in distribution substations under four optimized scheduling strategies were compared when a voltage warning mode existed in the distribution system. Voltage optimization results show that the single-mode adaptive optimized scheduling strategy, the multi-mode operation scheduling strategy, and the optimized scheduling strategy proposed in this invention can all effectively suppress voltage fluctuations on the high-voltage side of the main transformer in the distribution substations, significantly improving system voltage quality. Compared to the single-mode adaptive optimized scheduling strategy, the optimized strategy proposed in this invention further improves voltage quality by introducing a warning mode mechanism and using a piecewise function to construct the relationship between the optimization target weight coefficient and the voltage deviation, effectively reducing the risk of voltage exceeding limits during system operation. Existing multi-mode operation scheduling strategies only minimize voltage deviation as the optimization target when voltage fluctuations are large; therefore, the optimized scheduling strategy proposed in this invention outperforms its voltage optimization effect.

[0225] Table 3 shows the operating costs of different adaptive optimization scheduling strategies when the system is in voltage warning mode. The single-mode adaptive optimization scheduling strategy dynamically adjusts the weighting relationship between economy and voltage deviation. When the voltage deviation is small, it prioritizes system economy optimization. However, when the voltage deviation is large, to meet voltage correction requirements, it often sets the voltage deviation weight too high, resulting in excessive economic losses even with small voltage fluctuations. Therefore, compared to the segmented weight coefficient determination method proposed in this invention, it still suffers from significant economic sacrifice. While existing multi-mode optimization scheduling strategies determine the optimization objective by real-time monitoring of voltage values, they do not consider the coordinated optimization of economy and voltage quality under different operating modes. Therefore, their operating costs and voltage deviations are greater than those of the optimization scheduling strategy proposed in this invention.

[0226] Table 3

[0227] Strategy Weighting coefficient Operating cost / yuan Single-mode adaptive optimization scheduling Adaptive running state 75276 Existing multi-mode operation scheduling Fixed weights are used in different operating modes 76326 The optimized scheduling proposed in this invention Adaptive running state 75186

[0228] Simulation results show that when distribution substations enter the early warning operation mode, the optimized scheduling strategy proposed in this invention exhibits excellent multi-objective collaborative optimization capabilities. While ensuring the economical operation of the system, it effectively reduces the risk of voltage exceeding limits and significantly improves the system's safe and stable operation. Furthermore, the optimized scheduling strategy proposed in this invention converges in 10.94 seconds after 17 iterations, with a final convergence interval of 0.0002%, satisfying not only the time constraint of real-time distribution network scheduling but also the convergence accuracy requirements.

[0229] 3) Option 3: When the system enters the secondary voltage recovery mode, compare the fixed weight optimization scheduling strategy with the optimization scheduling strategy proposed in this invention.

[0230] When the load and DG experience larger power fluctuations, the transformer voltage will exceed the threshold, and the system will enter the secondary voltage recovery mode. Figure 11 To optimize the dynamic change process of the target weight coefficient when the power distribution system has a secondary voltage recovery mode, when the system voltage deviation is maintained within ±7%, the system does not need to start flexible loads for adjustment, and the weight coefficient of user satisfaction remains at 0. According to the optimization scheduling strategy proposed by this invention, the weight coefficient determination method under the secondary voltage recovery mode needs to be adopted to coordinate flexible loads to achieve rapid voltage adjustment while ensuring power supply reliability.

[0231] Figure 12 The voltage fluctuation distribution of each distribution area is shown under the fixed-weight optimization scheduling strategy. Figure 12 In (a), the voltage value of transformer area 32 in 12h has exceeded the stable operation threshold. Implementing a flexible load adjustment strategy based on demand-side response can effectively improve the system voltage deviation. The optimized voltage distribution of each transformer area is as follows: Figure 12 As shown in (b), the maximum voltage deviation of the distribution radio station group was controlled within ±7% within 24 hours.

[0232] Figure 13 The voltage fluctuation curves of each distribution area under the proposed optimized scheduling strategy are shown. Compared with the fixed weight coefficient optimized scheduling strategy, the multi-objective collaborative optimized scheduling strategy constructed in this invention controls the system operating voltage within ±5% by dynamically adjusting the target weights, thereby achieving the safe and stable operation of the distribution network.

[0233] Figure 14 The results of load regulation under different optimized scheduling strategies are presented. The adaptive optimized scheduling strategy constructed in this invention dynamically adjusts the optimization target weight coefficient by monitoring the voltage deviation amplitude in real time, and introduces flexible loads to participate in scheduling only when the voltage deviation exceeds a set threshold. Compared with the fixed weight coefficient optimized scheduling strategy, when the system has a secondary voltage recovery mode, the optimized scheduling strategy of this invention can reduce the total fluctuation amplitude of flexible loads by 92.22%, thereby improving the power consumption experience and power supply reliability for users in the distribution substation area.

[0234] The optimized scheduling strategy proposed in this invention reached a convergence state in 19.09 seconds after 13 iterations, with a convergence interval of 0.0099%, which meets the real-time control and convergence accuracy requirements of the system operation phase.

[0235] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for adaptive multi-objective cooperative scheduling of distribution radio areas based on flexible interconnection devices, characterized in that, Includes the following steps: Step (1) aims to minimize voltage fluctuations and flexible load fluctuations in the AC / DC power distribution system. An optimization objective function is constructed, with the power constraints of the SOP, power flow constraints of the system, system operation safety constraints, energy storage constraints, and flexible load constraints as constraints. The objective function is: min F=β1(t)f1'+β2(t)f2'+β3(t)f3' (1) In the formula, f1', f2', and f3' represent the normalized objective function values ​​of system operating cost, voltage deviation, and flexible load fluctuation, respectively; β1(t) is the weighting coefficient of system operating cost; β2(t) is the weighting coefficient of voltage deviation; and β3(t) is the weighting coefficient of flexible load fluctuation. Step (2) divides the transformer area operation mode into the following three typical states: steady-state operation mode, voltage warning mode and secondary voltage recovery mode; Dynamic weighting coefficients are constructed for steady-state operation mode, voltage warning mode, and secondary voltage recovery mode, respectively; Step (3) involves real-time detection of voltage fluctuations on the high-voltage side of the main transformer in the distribution area, calculation of dynamic weighting coefficients under the corresponding distribution area operation mode, solving the optimization objective function, and then controlling the distribution area according to the solution results.

2. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (1), the system operating cost f1 is composed of the electricity purchase cost C. pur Costs of wind and solar power curtailment (C) DG Power loss cost C loss And the cost of energy storage charging and discharging C ESS composition; f1=C pur +C DG +C loss +C ESS (2) In the formula, P(t) represents the electricity price at time t; P grid (t) represents the active power interaction value between the distribution area and the upper-level power grid at time t; c DG This indicates the penalty price for wind and solar power curtailment; P′ wt and P′ pv P represents the active power generated by the wind turbine and the photovoltaic generator, respectively. m,wt (t) and P m,pv (t) represents the actual active power utilized by the m-th wind turbine and the actual active power utilized by the m-th photovoltaic generator at time t, respectively, m = 1, 2, 3, ..., n; I ij (t) and R ij (t) represent the current and resistance on branch ij within the distribution area at time t, respectively, and φ1 represents the set of all branches within the distribution area; K ess This represents the unit charge / discharge cost coefficient for energy storage; r a y and y represent the depreciation rate and service life of energy storage, respectively; P h,ch and P h,dis These represent the charging and discharging power of the h-th energy storage device, respectively.

3. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (1), the specific calculation method for voltage deviation f2 is as follows: In the formula, U i (t) represents the high-voltage side voltage of the i-th distribution transformer at time t, n k This indicates the total number of stations.

4. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (1), the specific calculation method for the flexible load fluctuation f3 is as follows: In the formula, ΔP load This represents the change in flexible loads in the distribution network. and These are all decision variables for user-side scheduling, representing the planned power of the load that can be cut off, the increased load, and the decreased load that can be transferred at time t, respectively.

5. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (1), The power constraints of SOP are as follows: In the formula, and These represent the active power injected by SOP into nodes i and j during time period t, respectively. and These represent the reactive power injected by SOP into nodes i and j during time period t, respectively. and These represent the active power losses of the voltage source converters (VSCs) connected to node i and node j during time period t, respectively. and These represent the loss coefficients of SOP and VSC corresponding to node i and node j, respectively. and These represent the minimum and maximum reactive power injected by the VSC connected to node i, respectively. and These represent the minimum and maximum reactive power injected by the VSC connected to node j, respectively. and These represent the capacity limits of the VSC connected to node i and the VSC connected to node j, respectively. The specific power flow constraints of the system are: In the formula, P ij (t) and Q ij (t) represent the active and reactive power flowing through branch ij within the distribution area at time t, respectively; Ω i Let ψ represent the set of transformer substations starting from substation i. i R represents the set of transformer substations ending at substation i; li and X li These represent the resistance and reactance on branch line li of the transformer substation, respectively; I li (t) represents the current value flowing through branch li in the transformer area at time t; U l (t) and U i (t) represents the voltage amplitude of node l and node i in the distribution area at time t, respectively; and These represent the active and reactive power injected into the transformer area, respectively. The calculation expression is: In the formula, P i,DG (t) and Q i,DG (t) represent the active and reactive power injected by DG into transformer area i at time t, respectively; P i,SOP (t) and Q i,SOP (t) represent the active and reactive power injected by SOP at time t, respectively; and P represents the charging and discharging power of the i-th distribution station area at time t, respectively; i,Load_a (t) and Q i,Load (t) represents the active and reactive power required by the load at time t, respectively.

6. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (1), the specific system operation security constraints are as follows: In the formula, I ij (t) represent the current in branch ij of transformer area at time t; I ij,max U represents the maximum allowable current in branch ij; i (t) represents the effective voltage value of node i at time t; U i,min U represents the minimum allowable voltage at node i. i,max This represents the maximum allowable voltage at node i.

7. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (1), the energy storage constraints are specifically as follows: E h,min ≤E h (t)≤E h,max (19) In the formula, E h,max and E h,min E represents the upper and lower limits of the energy storage capacity of the h-th energy storage device, respectively; h The SOC value of the h-th energy storage device is expressed as follows: In the formula, E h (t) represents the SOC value of the energy stored in the h-th energy storage device at time t, E h (t-1) represents the SOC value of the energy storage device at the previous moment t-1 of the h-th energy storage device; This represents the energy storage charging power of the h-th energy storage device at time t; This represents the energy storage discharge power of the h-th energy storage device at time t; η ch and η dis Let represent the charge and discharge conversion efficiencies of the h-th energy storage device, respectively; Δt represents the scheduling time interval; to ensure the absolute continuity of the stored energy in time, the following also needs to be satisfied: AND h (0)=E h (24) (21) In the formula, E h (0) represents the initial SOC value of the h-th energy storage device at time 0; E h (24) represents the SOC value of the energy stored by the h-th energy storage device at the end of the cycle, i.e., at time 24. Energy storage charging and discharging constraints are: In the formula, and These represent the maximum energy storage charging and discharging power of the h-th energy storage device, respectively. and These represent the charging and discharging states of the energy stored in the h-th energy storage device.

8. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (1), the flexible load constraint is specifically as follows: When flexible loads participate in system scheduling, the power demand of the loads is: In the formula, P i,Load_a (t) represents the active power required by the load at time t; P i,Load (t) represents the active power required by the load before the transfer at time t; and These are all decision variables for user-side scheduling, representing the planned power of the load that can be cut off, the increased load, and the decreased load that can be transferred at time t, respectively. The constraints that transferable loads must satisfy are: In the formula, This represents the planned power of the transferable load added on the user side at time t; This represents the planned power of the transferable load reduced on the user side at time t; Indicates the upper limit of transferable load; The constraints that the load can be removed must satisfy are: In the formula, P in (t) represents the planned load that can be cut off on the user side at time t; and These represent the upper and lower limits of the removable load, respectively.

9. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (2): Stable operation mode: When the voltage deviation is kept within ±5% of the rated voltage, the system is in steady-state operation mode, and the main goal is to optimize economic efficiency. Voltage warning mode: When the voltage deviation is within ±5% to ±7% of the rated voltage, the system enters voltage warning mode, and voltage stability should be given priority at this time. Secondary voltage recovery mode: When the voltage deviation exceeds ±7% of the rated voltage, the system enters the secondary voltage recovery mode, at which time the primary goal is to quickly restore voltage stability.

10. The adaptive multi-objective cooperative optimization scheduling method for low-voltage distribution radio areas based on flexible interconnection devices according to claim 1, characterized in that, In step (2), the expression for the dynamic weighting coefficient in steady-state operation mode is: In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; ε1 and ε2 represent the linearization coefficients of the weighting coefficients, and ε1 represents the basic weight. ε2 represents the sensitivity coefficient to voltage deviation; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max This represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode; The expression for the dynamic weighting coefficient in voltage warning mode is: In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max U represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode; N This indicates the rated voltage value of the main transformer in the distribution substation area; The expression for the dynamic weighting coefficient in secondary voltage recovery mode: In the formula, β1(t) is the weighting coefficient of system operating cost, β2(t) is the weighting coefficient of voltage deviation, and β3(t) is the weighting coefficient of flexible load fluctuation; ε1 and ε2 represent the linearization coefficients of the weighting coefficients, and ε1 represents the basic weight. ε2 represents the sensitivity coefficient to voltage deviation; U max (t) represents the high-voltage side voltage of the transformer in the distribution substation with the largest voltage deviation during time period t; ΔU max This represents the maximum voltage deviation on the high-voltage side of the transformer in the distribution substation under steady-state operation mode.

Images