A power system reserve optimization method, system, processing device and storage medium considering reserve energy deliverability

Abstract

This application discloses a power system reserve optimization method considering reserve energy deliverability, comprising the following steps: obtaining an initial reliability cost model, and revising the initial reliability cost model according to the real-time response of reserve capacity to system disturbances to obtain an updated reliability cost model; linearizing the updated reliability cost model to obtain a mixed-integer linear programming model of reliability cost; obtaining an initial reserve optimization model for the power system, wherein the initial reserve optimization model includes at least the initial reliability cost model; updating the initial reserve optimization model to the initial reserve optimization model to obtain a reserve optimization model considering reserve deliverability; and optimizing the reserve dispatch of the power system using the reserve optimization model. The reserve optimization model obtained by this optimization method can yield a more deliverable reserve dispatch scheme.

Description

Technical Field

[0001] This application generally relates to the field of power system reserve optimization technology, and in particular to a power system reserve optimization method, system, processing device and storage medium that considers the availability of reserve energy. Background Technology

[0002] In power system dispatch and operation, to cope with potential unit failures, uncertainties in renewable energy output and load, the power system needs to rationally allocate a certain amount of reserve capacity. The allocation of reserve capacity is closely related to the economics and security of power system operation. When the reserve capacity is high, the system is able to cope with potential risks, and system security is guaranteed. However, in order to provide reserves, units may deviate from their most economical operating points, leading to a decrease in system economics. When the reserve capacity is low, the opportunity cost incurred by the system in reserving reserves is reduced, improving economics. However, the system is unable to cope with potential risks, and system security is reduced. Therefore, rationally allocating reserves is a key issue in achieving a balance between power system security and economics.

[0003] Traditional reserve optimization methods assume that demand or output remains constant on both the load and power supply sides within the same dispatch period, but varies in stages between different dispatch periods. In reality, this tiered dispatch satisfies the average output balance within the dispatch period, not the real-time power balance. Therefore, traditional reserve optimization models do not fully consider the reserve response process and confuse the concepts of reserve capacity and available reserve energy, assuming that reserve capacity is equivalent to available reserve energy. In the past, power systems had relatively low uncertainty, and due to limitations in computing power, treating the average output balance within the dispatch period as real-time power balance was reasonable and a necessary compromise for reserve optimization with long dispatch timescales. However, with the large-scale integration of renewable energy sources such as wind power, system uncertainty has increased, and power imbalances may occur more frequently within the same dispatch period. Although subsequent economic dispatch and automatic generation control reschedule processes can alleviate this problem to some extent, the reserve capacity that appears sufficient beforehand may not achieve the expected effect during deployment due to its inability to provide a step response. This leads to a greater risk of load shedding and wind curtailment than anticipated, making traditional tiered dispatch unsustainable.

[0004] Chinese patent document (CN113394789A) discloses an integrated power system dispatching method considering a high proportion of renewable energy access, addressing the issue of improving power system security and economy. This method constructs a multi-time granularity model, integrating the three components of UC, ED, and AGC (Power Providers), with the goal of minimizing the total operating cost of the power system. It then constructs constraints for this integrated power system dispatching model, obtaining an optimized model that satisfies all constraints. Robust optimization is performed on the uncertainty parameters in the constraints of the optimized model, and the final dispatching scheme is obtained by solving the robustly optimized model. This achieves the effect of improving power system security and economy. However, it does not disclose a solution for improving the deliverability of reserve energy in the power system. Summary of the Invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide a power system backup optimization method, system, processing device, and computer-readable storage medium that can improve the backup deliverability of power systems.

[0006] The specific technical solution is as follows:

[0007] First aspect

[0008] This application provides a power system backup optimization method considering backup energy deliverability, including the following steps:

[0009] Obtain an initial reliability cost model, and revise the initial reliability cost model according to the real-time response of the backup capacity to system disturbances, to obtain an updated reliability cost model;

[0010] The reliability cost update model is linearized to obtain a mixed-integer linear programming model for reliability cost.

[0011] Obtain an initial model for power system reserve optimization, wherein the initial model for reserve optimization includes at least: an initial model for reliability cost;

[0012] The mixed-integer linear programming model is updated to the alternative optimization initial model to obtain an alternative optimization model that considers alternative deliverability.

[0013] The backup scheduling of the power system is optimized using the aforementioned backup optimization model.

[0014] As a further limitation of this application, the initial model for obtaining reliability cost includes at least: an initial model for expected power shortage and an initial model for expected wind curtailment.

[0015] As a further limitation of this application, the step of correcting the initial reliability cost model based on the real-time response of backup capacity to system disturbances, and obtaining the updated reliability cost model, includes the following steps:

[0016] Obtain the initial model of expected power shortage, and revise the initial model of expected power shortage based on the real-time response of increasing reserve capacity to system disturbances, to obtain the updated model of expected power shortage;

[0017] Obtain the initial model of expected wind curtailment, and revise the initial model of expected wind curtailment based on the real-time response to system disturbances by reducing reserve capacity, to obtain the updated model of expected wind curtailment.

[0018] The reliability cost update model includes: the expected power shortage update model and the expected wind curtailment update model.

[0019] As a further limitation of this application, the linearization process of the reliability cost update model to obtain a mixed-integer linear programming model of reliability cost includes the following steps:

[0020] Linearizing the expected power shortage update model yields a mixed-integer linear programming model for the expected power shortage.

[0021] The wind curtailment expectation update model is linearized to obtain a mixed-integer linear programming model for the wind curtailment expectation.

[0022] As a further limitation of this application, the steps for obtaining the initial model for power system reserve optimization include:

[0023] Obtain the objective function of the backup optimization initial model, wherein the objective function is to minimize the sum of operating cost, backup cost and reliability cost;

[0024] Obtain the constraints of the initial standby optimization model, including at least: generator output constraints, logic variable constraints, power balance constraints, unit ramping constraints, unit minimum start-up and shutdown time constraints, standby adjustment constraints, and standby reduction constraints.

[0025] Second aspect

[0026] This application provides a power system backup optimization system that considers the availability of backup energy, including a first update module. The first update module is configured to obtain an initial reliability cost model and, based on whether the backup capacity is sufficient to cope with system disturbances, correct the initial reliability cost model to obtain an updated reliability cost model.

[0027] A linearization module is configured to linearize the reliability cost update model to obtain a mixed-integer linear programming model of reliability cost.

[0028] The acquisition module is configured to acquire an initial backup optimization model of the power system, the initial backup optimization model including at least: an initial reliability cost model;

[0029] The second update module is configured to update the mixed-integer linear programming model to the backup optimization initial model to obtain a backup optimization model that takes into account backup deliverability.

[0030] Third aspect

[0031] This application provides a processing apparatus, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the power system backup optimization method that takes into account backup energy deliverability as described above.

[0032] Fourth aspect

[0033] This application provides a computer-readable storage medium having a computer program that, when executed by a processor, implements the steps of the power system backup optimization method considering backup energy deliverability as described above.

[0034] The beneficial effects of this application are:

[0035] (1) This invention differs from traditional reserve optimization methods. Based on the optimization of reserve capacity, it analyzes the deliverability of reserve energy in detail and establishes a new reliability cost model based on the actual deliverable reserve energy, thereby making a more accurate assessment of the wind curtailment and load shedding risks faced by the power system.

[0036] (2) The present invention further linearizes the calculation of reliability cost by using piecewise linearization and the big M method to transform the model into a mixed integer linear programming model that can be solved by directly calling commercial solvers, thereby improving the computational efficiency of the model and making the model more practical.

[0037] (3) This invention establishes a backup optimization model that considers the deliverability of backup energy. With the objective function of minimizing the sum of operating cost, backup cost and reliability cost, under the constraints of power balance, unit operation and unit backup, the uncertainty of wind power, load and equipment failure is considered, and then a more practical and reliable backup optimization scheme is determined. Attached Figure Description

[0038] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0039] Figure 1A flowchart of a power system backup optimization method considering backup energy deliverability provided for embodiments of this application;

[0040] Figure 2 Assuming the backup energy is output in a stepped manner, when the system disturbance exceeds the increased backup capacity, the output power curve of the backup capacity after event s;

[0041] Figure 3 Assuming the backup energy is output in a stepped manner, the output power curve of the backup capacity after event s occurs when the increase in backup capacity exceeds the system disturbance.

[0042] Figure 4 Assuming the backup energy is output gradually, the output power curve of the backup capacity after event s occurs when the system disturbance exceeds the increased backup capacity.

[0043] Figure 5 Assuming the backup energy is output gradually, the output power curve of the backup capacity after event s occurs when the increase in backup capacity exceeds the system disturbance.

[0044] Figure 6 Assuming the backup energy is output in a stepped manner, when the system disturbance exceeds the reduction in backup capacity, the output power curve of the backup capacity after event s occurs;

[0045] Figure 7 Assuming the backup energy is output in a stepped manner, the output power curve of the backup capacity after event s occurs when the down-adjustment of the backup capacity exceeds the system disturbance.

[0046] Figure 8 Assuming the backup energy is output gradually, the output power curve of the backup capacity after event s occurs when the system disturbance exceeds the reduction in backup capacity;

[0047] Figure 9 Assuming the backup energy is output gradually, the output power curve of the backup capacity after event s occurs when the down-adjustment of the backup capacity exceeds the system disturbance.

[0048] Figure 10 The curve showing the change in expected power shortage as reserve capacity is increased;

[0049] Figure 11 The curve showing the change in expected wind curtailment as reserve capacity is reduced;

[0050] Figure 12 When the increase in reserve capacity exceeds the system disturbance, the first unit will serve as a backup for external power output during the first time period.

[0051] Figure 13When the system disturbance exceeds the increased reserve capacity, the second unit will serve as a backup for external power output during the second time period. Detailed Implementation

[0052] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0053] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0054] Example 1

[0055] Please refer to Figure 1 The flowchart of a power system backup optimization method considering backup energy deliverability provided in this embodiment includes the following steps:

[0056] S100: Obtain the initial reliability cost model, and revise the initial reliability cost model according to the real-time response of the backup capacity to system disturbances, to obtain the updated reliability cost model;

[0057] S200: Linearize the reliability cost update model to obtain a mixed-integer linear programming model for reliability cost;

[0058] S300: Obtain the initial model for power system reserve optimization, wherein the initial model for reserve optimization includes at least: an initial model for reliability cost;

[0059] S400: Update the mixed-integer linear programming model to the backup optimization initial model to obtain a backup optimization model that considers backup deliverability;

[0060] S500: Optimize the standby scheduling of the power system using the aforementioned standby optimization model.

[0061] Before elaborating on the model optimization process, the following concepts need to be clarified:

[0062] Expected Energy Not Supplied (EENS) is the expected power shortage that the power system will supply to external systems when an uncertain event (increased demand) occurs on the load side.

[0063] Expected Wind Energy Spillage (EWS) is the expected amount of excess power that the power system will supply to external systems after an uncertain event (reduction in demand) occurs on the load side.

[0064] Both of these variables will affect the economics of power system operation.

[0065] The specific implementation process of the above method is as follows:

[0066] I. Obtaining the updated reliability cost model from the initial reliability cost model:

[0067] (I) Obtain the initial model of expected power shortage and expected wind curtailment. The traditional expressions for EENS and EWS are as follows:

[0068]

[0069]

[0070]

[0071]

[0072] In equations (1) and (2), S up t p represents the set of all events that could cause a power system load loss within time period t; up s,t ΔC represents the probability of event s occurring within time period t; up s,t b represents the power deficit caused by event s occurring within time period t; up s,t This represents a 0 / 1 variable indicating whether an event s occurring within time period t will cause a load shedding. When the power deficit exceeds the system's available reserve capacity, b... up s,t It is 1 if it is true, otherwise it is 0.

[0073] In equations (3) and (4), S dn t p represents the set of all events that could potentially cause wind curtailment within time period t; dn s,t ΔC represents the probability of event s occurring within time period t; dn s,t This indicates excess power caused by event s occurring within time period t; b dn s,t This represents a 0 / 1 variable indicating whether an event s occurring within time period t will cause wind curtailment. When the excess power exceeds the system's available reserve capacity, b... dn s,t It is 1 if it is true, otherwise it is 0.

[0074] The initial reliability cost model includes at least: the expected power shortage, i.e., the EENS initial model, and the expected wind curtailment, i.e., the EWS initial model. In the initial reliability cost model, EENS is shown in Equation (1), and EWS is shown in Equation (3).

[0075] (ii) The above model takes into account both the real-time nature of increasing reserve capacity to respond to system disturbances and the real-time nature of decreasing reserve capacity to respond to system disturbances:

[0076] 0 / 1 variable b in traditional EENS expressions up s,t To ensure that load shedding has only two outcomes, load shedding will occur if the system disturbance exceeds the total reserve requirement; otherwise, it will not. These two outcomes can be determined by the attached... Figure 2 and 3 This is presented more intuitively. (Attached) Figure 2 and 3 In the middle, the response trajectory for adjusting the standby level is indicated by a bold black line. (Attached) Figure 2 This indicates a situation where the system disturbance exceeds the increased reserve amount; the shaded area represents EENS. Figure 3 This indicates that increasing the total reserve exceeds the system disturbance will not cause a load shedding. It can be seen that the traditional EENS model assumes that increasing the reserve can be a step change; however, considering actual operating conditions, the response trajectory of increasing the reserve depends on its actual ramp-up process, as shown in the attached figure. Figure 4 and 5 As shown. (Attached) Figure 4 and 5 The area of ​​the shaded region in the diagram represents the actual EENS caused. (See attached diagram) Figure 2 and attached Figure 4 The comparison shows that when increasing reserve capacity is insufficient, considering the ramp-up process of increasing reserve will result in more ENS than assuming that increasing reserve will lead to a step response; (Comparison with attached...) Figure 3 and attached Figure 5 It can be seen that even if the reserve capacity is sufficient, considering the ramp-up process of reserve increases will still cause EENS (Energy Efficiency and Resilience), rather than completely eliminating load loss. In summary, ignoring the ramp-up process of reserve increases overestimates the energy deliverable from reserve increases, thus underestimating the risk of load loss.

[0077] Based on the above analysis, EENS can be calculated in two cases, depending on whether increasing the reserve capacity is sufficient to cope with system disturbances. To distinguish between these two cases, a 0 / 1 variable b is introduced. up_ins s,t and b up_suf s,t When the increased reserve capacity is insufficient, b up _ins s,t Take 1 and b up_sufs,t If the value is 0, then the corresponding ENS is denoted as ENS. ins s,t When the reserve capacity is sufficient, b up_suf s,t Take 1 and b up_ins s,t If the value is 0, then the corresponding ENS is denoted as ENS. suf s,t Therefore, EENS can be expressed as follows:

[0078]

[0079]

[0080]

[0081] EENS ins s,t Corresponding Appendix Figure 4 The area of ​​the shaded region in the equation is relatively simple to calculate, as shown in equation (8). EENS suf s,t Corresponding Appendix Figure 5 The shaded area in the figure represents the area where the system has already achieved power balance before the end of the scheduling period, and no further load shedding is required. Assuming the system has achieved power balance at time τ1, τ1 is calculated as shown in equation (9), EENS suf s,t The calculation is shown in equation (10).

[0082]

[0083]

[0084]

[0085] Ultimately, EENS can be represented as:

[0086]

[0087] Similarly, the 0 / 1 variable b in the traditional EWS expression dn s,t To ensure wind curtailment has only two outcomes: it will occur when system disturbances exceed the total reserve reduction; otherwise, it will not. These two outcomes can be determined by the attached... Figure 6 and 7 This is presented more intuitively. (See appendix) Figure 6 and 7 In the middle, the reduced reserve response trajectory is indicated by a bold black line, with an appendix. Figure 6 This indicates a situation where the system disturbance exceeds the total reserve reduction; the shaded area is the EWS (Extended System Warp / Shaded Area). Figure 7 This indicates that reducing the total reserve amount beyond the system disturbance will not cause wind curtailment. It can be seen that traditional models assume reserve reduction can be abrupt; however, considering actual operation, the response trajectory of reserve reduction depends on its actual ramp-up process, as shown in the attached figure. Figure 8 and 9 As shown. In the appendix Figure 8 and 9 The shaded area in the diagram represents the actual EWS caused. (See attached image.) Figure 8 and attached Figure 9 The comparison shows that when reserve capacity is insufficient, considering the ramp-up process of reserve reduction results in more EWS (Extreme Strike Loss) than assuming that reserve reduction can achieve a step response; (Comparison with attached...) Figure 8 and attached Figure 9 It can be seen that even if the reserve capacity reduction is sufficient, considering the ramp-up process of reserve reduction will still result in wind curtailment (EWS), rather than completely eliminating wind curtailment. In summary, ignoring the ramp-up process of reserve reduction overestimates the energy deliverable from the reduced reserve, thus underestimating the risk of wind curtailment.

[0088] Based on the above analysis, the calculation of EWS can be divided into two cases based on whether reducing the reserve capacity is sufficient to cope with system disturbances. To distinguish between these two cases, a 0 / 1 variable b is introduced. dn_ins s,t and b dn_suf s,t When the reserve capacity is insufficient, b dn_ins s,t Take 1 and b dn_suf s,t When the value is 0, the corresponding EWS is denoted as EWS. ins s,t When the reserve capacity is sufficient, b dn_suf s,t Take 1 and b dn_ins s,t When the value is 0, the corresponding EWS is denoted as EWS. suf s,t Therefore, EWS can be expressed as follows:

[0089]

[0090]

[0091]

[0092] EWS ins s,t Corresponding Appendix Figure 8 The area of ​​the shaded region in EWS is relatively simple to calculate, as shown in equation (15). suf s,t Corresponding Appendix Figure 9 The area of ​​the shaded region in the equation indicates that the system has already achieved power balance before the end of the scheduling period, and there is no need to continue curtailing wind power. Assuming the system has achieved power balance at time τ2, τ2 is calculated as shown in equation (16). suf s,t The calculation is shown in equation (17).

[0093]

[0094]

[0095]

[0096] Ultimately, EWS can be represented as:

[0097]

[0098] (iii) Obtain the expected power shortage update model and the expected wind curtailment update model according to equations (11) and (18), and obtain the reliability cost update model according to the above two models.

[0099] Second, the reliability cost update model is linearized to obtain a mixed-integer linear programming model for reliability cost, specifically including:

[0100] (a) Linearizing the expected power shortage update model yields a mixed-integer linear programming model for the expected power shortage:

[0101] The linearization of EENS can be divided into two cases. For the case where the increased reserve capacity is insufficient, an auxiliary variable z can be introduced. up s,t The linearization results are shown in equations (19) and (20):

[0102]

[0103]

[0104] If the increased reserve capacity is sufficient, EENS suf s,t The relationship between this and increasing reserve capacity can be found in the appendix. Figure 10 Visually intuitive.

[0105] As attached Figure 10 As shown, with the increase in reserve capacity, EENS suf s,tThe trajectory of the change follows an inverse proportional function curve, therefore piecewise linearization can be performed, approximating the inverse proportional function curve with the red line segment. Taking integer multiples of the disturbance as segmentation points, and calculating the function value at each segmentation point, we can then determine the slope and intercept of each line segment. The piecewise linearized EENS... suf s,t The following constraints must be satisfied:

[0106]

[0107]

[0108]

[0109]

[0110]

[0111] In the formula, b up_ss m,s,t This indicates a decision to increase the reserve capacity SSR. up t Whether the variable falls within the 0 / 1 range of the m-th disturbance interval, when the reserve capacity SSR is increased. up t When it falls within the m-th segment of the disturbance range, b up_ss m,s,t Select 1 otherwise select 0. EENS suf m,s,t Indicates EENS at each segment point suf s,t The value of a up_suf m,s,t means EENS suf s,t The slope, b, after linearization of the m-th segment of the disturbance interval up_suf m,s,t means EENS suf s,t The intercept after linearization of the m-th segment of the disturbance range.

[0112]

[0113]

[0114] Equation (21) still contains the product of 0 / 1 variables and continuous variables, requiring further linearization. First, for the 0 / 1 variable b... up_ss m,s,t Equation (22) is only a logical expression and requires further processing. Therefore, a new 0 / 1 variable b is introduced. up_s m,s,t and b up_sssm,s,t And satisfy constraints (26) and (27) to replace equation (22). In equation (26), a sufficiently large positive number can ensure that when SSR up t <mΔC up s,t At that time, b up_s m,s,t Take 1, otherwise take 0. Equation (27) introduces a 0 / 1 variable b. up_sss m,s,t It can be ensured that when mΔC up s,t ≤SSR up t <(m+1)ΔC up s,t At that time, b up_ss m,s,t Select 1 if the value is 1, otherwise select 0.

[0115] 0 / 1 variable b up_ss m,s,t After being expressed by constraints (26) and (27), the product of the 0 / 1 variables and the continuous variables in equation (21) can be linearized using the Big M method, and equation (21) can be replaced by equations (28)-(31).

[0116]

[0117]

[0118]

[0119]

[0120] (ii) Linearizing the wind curtailment expectation update model to obtain a mixed-integer linear programming model for the wind curtailment expectation:

[0121] Linearization of EWS can be divided into two cases. For the case where the reduced reserve capacity is insufficient, an auxiliary variable z can be introduced. dn s,t The linearization results are shown in equations (32) and (33):

[0122]

[0123]

[0124] If the reduction in standby capacity is sufficient, EWS suf s,t The relationship between this and reducing reserve capacity can be found in the appendix. Figure 11 Visually intuitive.

[0125] As attached Figure 11As shown, with the increase in the reduction of reserve capacity, EWS suf s,t The trajectory of the change follows an inverse proportional function curve, therefore piecewise linearization can be performed, approximating the inverse proportional function curve with the red line segment. Integer multiples of the disturbance are taken as segmentation points, and the function value at each segmentation point is calculated. This allows us to determine the slope and intercept of each line segment, resulting in the piecewise linearized EWS. suf s,t The following constraints must be satisfied:

[0126]

[0127]

[0128]

[0129]

[0130]

[0131] In the formula, b dn_ss m,s,t This indicates a decision to reduce the reserve capacity SSR. dn t Whether the variable falls within the 0 / 1 range of the m-th disturbance interval, the current reserve capacity SSR is adjusted. dn t When it falls within the m-th segment of the disturbance range, b dn_ss m,s,t Select 1 otherwise select 0. EWS suf m,s,t Indicates the EWS at each segment point suf s,t The value of a dn_suf m,s,t EWS suf s,t The slope, b, after linearization of the m-th segment of the disturbance interval dn_suf m,s,t EWS suf s,t The intercept after linearization of the m-th segment of the disturbance range.

[0132]

[0133]

[0134] Equation (34) still contains the product of 0 / 1 variables and continuous variables, requiring further linearization. First, for the 0 / 1 variable b... dn_ss m,s,t Equation (35) is only a logical expression and requires further processing. Therefore, a new 0 / 1 variable b is introduced.dn_s m,s,t and b dn_sss m,s,t And satisfy constraints (39) and (40) to replace equation (35), in equation (39), M is a sufficiently large positive number that can ensure that when SSR dn t <mΔC dn s,t At that time, b dn_s m,s,t Take 1, otherwise take 0. Equation (40) introduces a 0 / 1 variable b. dn_sss m,s,t It can be ensured that when mΔC dn s,t ≤SSR dn t <(m+1)ΔC dn s,t At that time, b dn_ss m,s,t Select 1 if the value is 1, otherwise select 0.

[0135] 0 / 1 variable b dn_ss m,s,t After being expressed by constraints (39) and (40), the product of the 0 / 1 variables and the continuous variables in equation (34) can be linearized using the Big M method, and equation (34) can be replaced by equation (41)-(44).

[0136]

[0137]

[0138]

[0139]

[0140] 3. Update the mixed-integer linear programming model to the alternative initial optimization model to obtain the alternative optimization model considering alternative deliverability:

[0141] (a) Obtaining the objective function of the alternative initial optimization model:

[0142] The objective function of the model is to minimize the operating cost, backup cost, and reliability cost.

[0143]

[0144] In the formula, g / N G It is a set and index of generators, t / N T It optimizes the collection and index of time periods; C NL g For the no-load cost of generator set g, C LVg For the fuel consumption cost of generator set g, C up g and C dn g These represent the start-up and shutdown costs of generator set g and u, respectively. g,t b up g,t and b dn g,t The variable is 0 / 1, with a value of 1 indicating that generator unit g is in the running, starting, or shut-down state during time period t, respectively; otherwise, it is 0. g,t π represents the electrical energy output of generator set g during time period t; up g and π dn g Adjusting the reserve cost upwards and downwards for generator set g respectively; R up g,t and R dn g,t These represent the increase and decrease of the reserve capacity of generator unit g during time period t, respectively. VOLL is the off-load penalty price, and VOLW is the wind curtailment penalty price. In equation (45), the expressions for EENS are (28)-(31), and the expressions for EWS are (41)-(44).

[0145] (ii) Obtaining the constraints of the backup initial optimization model:

[0146] The constraints of the model are as follows:

[0147] 1. Generator output constraints:

[0148]

[0149]

[0150]

[0151] In the formula, P g,t P represents the output power of generator set g at the end of time period t. max g P min g P represents the upper and lower limits of the output power of generator set g. 0 g is the generator set output power at the initial moment; Equation (46) is the upper and lower limit constraint of the generator set output power, and Equations (47) and (48) are the power output constraints of the generator set in each time period.

[0152] 2. Logical variable constraints:

[0153]

[0154]

[0155] In the formula, u 0 g This is a binary variable representing the initial operating state of generator set g (1 for running, 0 for shutting down).

[0156] 3. Power balance constraints:

[0157]

[0158] In the formula, P L t Let P be the load at the end of time period t. WT t Let t be the wind power output at the end of time period t.

[0159] 4. Unit ramp-up constraints:

[0160]

[0161]

[0162] In the formula, UR g and DR g These represent the climb rate and descent rate of generator set g, respectively.

[0163] Initial ramp-up constraints for the unit:

[0164]

[0165]

[0166] In the formula, IC g This indicates the initial operating state of generator set g, and its value represents the operating time; T on g and T off g These represent the minimum operating time and minimum downtime of generator set g, respectively.

[0167] 5. Minimum start-up and shutdown time constraints for the unit:

[0168]

[0169]

[0170]

[0171]

[0172] Equations (56) and (57) are the minimum running time constraint and the initial minimum running time constraint, respectively; Equations (58) and (59) are the minimum downtime constraint and the initial minimum downtime constraint, respectively.

[0173] 6. Increase standby constraints:

[0174]

[0175]

[0176]

[0177]

[0178] In the formula, SSR up t This represents the total upward reserve of the system. Equations (61) and (62) respectively represent the constraints on the actual available upward reserve capacity, considering the impact of ramp-up constraints on the reserve response process.

[0179] 7. Lower the standby constraint:

[0180]

[0181]

[0182]

[0183]

[0184] In the formula, SSR dn t This represents the total down-limit reserve of the system. Equations (65) and (66) respectively represent the actual available down-limit reserve capacity constraints considering the impact of ramp-up constraints on the reserve response process.

[0185] The IEEE-RTS 26-unit system is used for the case study analysis. The price for spinning reserve provided by all units in the system is set to 10% of the fuel cost; the wind power output follows a Gaussian distribution, with its mean set to 20% of the system load and its standard deviation set to 15% of the mean; the standard deviation of the load forecast error is set to 3% of the system load; VOLL is set to $1000 / MWh and VOLW is set to $500 / MWh.

[0186] The proposed model is implemented using the GAMS platform and solved using the commercial solver CPLEX, achieving a convergence accuracy of 0.1%. The computer configuration is an Intel Core i5-4460 series processor with a clock speed of 3.2GHz and 8GB of memory.

[0187] This application provides numerical examples to compare the two schemes.

[0188] Option 1: Disregarding the availability of backup energy;

[0189] Option 2: Consider the availability of backup energy.

[0190] The costs for the two options are shown in Table 1:

[0191] Table 1. Cost Comparison of the Two Options:

[0192]

[0193] Observing Table 1, it can be seen that compared to Scheme 1, which does not consider the availability of reserve energy, Scheme 2, which considers the availability of reserve energy, has increased total cost, operating cost, and start-up and shutdown costs. The main reason is that, considering the availability of reserve energy, the generator units are not at their optimal operating point to provide more reliable backup, thus reducing system economics. Furthermore, considering the availability of reserve energy significantly increases backup costs and reliability costs; backup costs almost double. However, even with more reserves, the costs of load shedding and wind curtailment still increase, with load shedding costs increasing by approximately 5.98 times, and wind curtailment costs no longer being zero. The main reason for the change in wind curtailment costs is that Scheme 1 assumes that if the reserve capacity is sufficiently reduced, the power system will not face any wind curtailment risk, thus the cost of wind curtailment is zero. However, in reality, although reducing the reserve capacity seems sufficient, the system still faces the risk of wind curtailment because the backup cannot respond instantly. The cost comparison of the two schemes shows that considering the availability of reserve energy leads to a decrease in system economics, but it provides a more accurate assessment of wind curtailment and load shedding risks, better ensuring the safety of the power system.

[0194] To illustrate the impact of reserve capacity adjustments on the actual deliverable energy, the scenario of the first unit being shut down in the first period and the net load fluctuation being the most moderate is used as an example. At this time, the system load loss is 371.231MW, while the system can provide 445.471MW of reserve capacity. Clearly, the system's reserve capacity adjustment is sufficient to cope with system fluctuations. However, because the reserve cannot provide a step response, it will still cause issues such as… Figure 12 The shaded area represents the load shedding. However, if the availability of backup energy is not considered, it is mistakenly assumed that there will be no load shedding, thus underestimating the risk of system load shedding.

[0195] To address situations where increased reserve capacity is insufficient to cope with system fluctuations, consider the scenario where the second generating unit shuts down during the second period and the net load fluctuates most drastically. In this case, the system load loss is 502.481MW, while the system can provide 445.477MW of increased reserve capacity. Clearly, the increased reserve capacity is insufficient to handle system fluctuations. However, if we disregard the impact of the ramp-up process on the actual available reserve energy, the system power shortage would be considered to be (502.481MW - 445.477MW) * 1h = 57.004MWh. But if we consider the impact of the ramp-up process on the actual available reserve energy, the actual power shortage is 445.477MW / 2 * 1h + 57.004MWh = 279.7425MWh. Figure 13 The area of ​​the shaded region is shown in the figure. It is evident that neglecting the availability of backup energy significantly underestimates the risk of system failure.

[0196] Example 2

[0197] This embodiment provides a power system backup optimization system that considers the availability of backup energy, including:

[0198] The first update module is configured to obtain an initial reliability cost model and, based on whether the backup capacity is sufficient to cope with system disturbances, correct the initial reliability cost model to obtain an updated reliability cost model.

[0199] A linearization module is configured to linearize the reliability cost update model to obtain a mixed-integer linear programming model of reliability cost.

[0200] The acquisition module is configured to acquire an initial backup optimization model of the power system, the initial backup optimization model including at least: an initial reliability cost model;

[0201] The second update module is configured to update the mixed-integer linear programming model to the backup optimization initial model to obtain a backup optimization model that takes into account backup deliverability.

[0202] Example 3

[0203] This embodiment provides a processing apparatus, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the power system backup optimization method that considers backup energy deliverability as described above.

[0204] Example 4

[0205] This embodiment provides a computer-readable storage medium having a computer program. When the computer program is executed by a processor, it implements the steps of the power system backup optimization method considering backup energy deliverability as described above.

[0206] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for reserve optimization of a power system considering deliverability of reserve energy, characterized in that, The method comprises the following steps: obtaining an initial reliability cost model, and correcting the initial reliability cost model according to real-time performance of reserve capacity in response to system disturbance to obtain an updated reliability cost model; linearizing the updated reliability cost model to obtain a mixed integer linear programming model of the reliability cost; obtaining an initial reserve optimization model of the power system, wherein the initial reserve optimization model at least comprises the initial reliability cost model; updating the mixed integer linear programming model to the initial reserve optimization model to obtain a reserve optimization model considering reserve deliverability; optimizing reserve scheduling of the power system based on the reserve optimization model; the initial reliability cost model at least comprises an initial expected power supply shortage model and an initial expected wind curtailment model; the step of correcting the initial reliability cost model according to real-time performance of reserve capacity in response to system disturbance to obtain an updated reliability cost model comprises the following steps: obtaining an initial expected power supply shortage model, and correcting the initial expected power supply shortage model according to real-time performance of up-regulated reserve capacity in response to system disturbance to obtain an updated expected power supply shortage model; obtaining an initial expected wind curtailment model, and correcting the initial expected wind curtailment model according to real-time performance of down-regulated reserve capacity in response to system disturbance to obtain an updated expected wind curtailment model; the updated reliability cost model comprises the updated expected power supply shortage model and the updated expected wind curtailment model; when correcting the initial expected wind curtailment model according to real-time performance of down-regulated reserve capacity in response to system disturbance, gradual output characteristics of down-regulated reserve caused by ramping constraints need to be considered; the correction is performed in two cases: when down-regulated reserve capacity is insufficient to respond to system disturbance, more expected wind curtailment than the step response scenario after considering the ramping process is calculated; when down-regulated reserve capacity is sufficient to respond to system disturbance, non-zero expected wind curtailment caused by the ramping process is calculated to avoid overestimating deliverable energy of down-regulated reserve and underestimating wind curtailment risk due to neglecting the ramping process, thereby obtaining the updated expected wind curtailment model.

2. The method for reserve optimization of a power system considering deliverability of reserve energy according to claim 1, characterized in that, the step of linearizing the updated reliability cost model to obtain a mixed integer linear programming model of the reliability cost comprises the following steps: linearizing the updated expected power supply shortage model to obtain a mixed integer linear programming model of the expected power supply shortage; linearizing the updated expected wind curtailment model to obtain a mixed integer linear programming model of the expected wind curtailment.

3. The method for reserve optimization of a power system considering deliverability of reserve energy according to claim 2, characterized in that, the step of obtaining the initial reserve optimization model of the power system comprises: obtaining an objective function of the initial reserve optimization model, wherein the objective function is to minimize the sum of operating cost, reserve cost and reliability cost; obtaining constraint conditions of the initial reserve optimization model, wherein the constraint conditions at least comprise generator output constraints, logical variable constraints, power balance constraints, unit ramping constraints, unit minimum start-stop time constraints, up-regulated reserve constraints and down-regulated reserve constraints.

4. A power system reserve optimization system considering reserve energy deliverability, characterized by, The system implements the steps of the power system reserve optimization method considering reserve deliverability according to any one of claims 1 to 3, and comprises: a first updating module configured to obtain an initial reliability cost model, and correct the initial reliability cost model according to whether reserve capacity is sufficient to respond to system disturbance to obtain an updated reliability cost model. a linearization module, configured to linearize a reliability cost updating model to obtain a mixed integer linear programming model of reliability cost; a collecting module, configured to obtain an initial model of power system reserve optimization, the initial model of reserve optimization at least comprising an initial model of reliability cost; a second updating module, configured to update the mixed integer linear programming model to the initial model of reserve optimization to obtain a reserve optimization model considering reserve deliverability.

5. A processing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the steps of the power system reserve optimization method considering reserve energy deliverability as claimed in any one of claims 1 to 3.

6. A computer readable storage medium having a computer program, characterized in that The computer program is executed by the processor to implement the steps of the power system reserve optimization method considering reserve energy deliverability as claimed in any one of claims 1 to 3.

Images