AGC total power allocation strategy considering vibration zones in the FM market

Abstract

The present invention discloses an AGC total power allocation strategy that takes vibration zones into consideration in a frequency regulation market, including: step 1, obtaining a command value based on data collected in real time from the power grid and in accordance with the total regulation instruction strategy; step 2, constructing a unit model, obtaining the unit operating zone, and conceptually defining the number of crossings, complete crossing mileage, and time spent in the vibration zone involved in the constructed model; step 3, constructing multiple objective functions and constraints; step 4, defining the importance of each objective function, and obtaining the coefficients of each objective function used in the weighted sum method, thereby converting the multi-objective problem into a single-objective problem; and step 5, solving the constructed model to obtain the final allocation strategy. The allocation strategy of the present invention reduces the number of crossings and the time the unit stays in the vibration zone, reduces the loss of the hydropower AGC unit, reduces the total cost of social operation, and fully mobilizes the enthusiasm of each unit with different frequency regulation mileage ranking prices.

Description

Technical Field

[0001] An embodiment of the present invention relates to an AGC total power allocation strategy considering vibration zones in a frequency modulation market. Background Art

[0002] The dispatching center often issues adjustment instructions to the direct-controlled units based on work experience or the adjustable capacity of the units. The power plant must cooperate with the AGC frequency regulation free of charge, distribute the total adjustment power according to the adjustable capacity of the units, and obtain power adjustment instructions for each AGC unit. This dispatching method is simple and easy for the dispatching center, but there will be situations where the direct-controlled AGC units cannot complete the instructions well and are instead subject to assessment by the power grid.

[0003] In the frequency regulation auxiliary service market, units submit applications based on their actual conditions, and after bidding and sorting, they are cleared in order to obtain the AGC units that will participate in regulation in the future period. In the subsequent real-time allocation, each time the total regional regulation power is generated, the regional total regulation power is allocated to each AGC unit in sequence according to the priority of the sorting price until the sum of the power allocated to each unit equals the total regional regulation power. Allocating regulation instructions in order of ranking price can quickly obtain the final power allocation result, but this allocation strategy ignores the actual operating conditions of the units when allocating power. For hydropower units, the most important thing is that it ignores the existence of vibration zones. The strategy of allocating regulation instructions only based on ranking price will cause some hydropower units to frequently cross the vibration zone or remain in the vibration zone for a long time. Summary of the Invention

[0004] The purpose of the present invention is to provide an AGC total power allocation strategy considering vibration zones in a frequency modulation market.

[0005] To achieve the above-mentioned purpose, the present invention adopts the following technical solutions: A total power allocation strategy of AGC considering the vibration zone in the frequency modulation market, see Figure 1 , including the following steps,

[0006] Step 1: Based on the real-time data collected from the power grid and in accordance with the overall regulation instruction strategy, a command value is obtained. Traditional PI control can be selected as the controller, or various other types of controllers can be selected.

[0007] Step 2: Construct the unit model and divide the output operating range of the hydropower unit into different areas according to the difference between the feasible area and the vibration area. The areas are divided and numbered from bottom to top to obtain the unit operating area (see Figure 2 ), and conceptually define the number of spans, complete span mileage, and time spent passing through the vibration zone involved in the constructed model.

[0008] Assume that the first area is the operating area, followed by a prohibited area, and finally ends with the mth operating area. Therefore, a unit has a total of m operating areas and (m-1) prohibited areas. ijmax and x ijmin They represent the upper and lower limits of the jth feasible region of the i-th unit, respectively. Since the jth vibration region of the i-th unit is adjacent to the jth feasible region of the i-th unit, x i(j+1)min and x ijmax They represent the upper and lower limits of the jth vibration zone of the i-th unit. ij =x i(j+1)min -x ijmax Indicates the value of the jth vibration zone of the i-th unit, which can be correspondingly expressed by Q ij =x ijmax -x ijmin Indicates the numerical value of the jth feasible area of ​​the i-th unit. If x i1max =x i1min It means that the first area of ​​power from bottom to top is the vibration area. If x immax =x immin This means that the last area of ​​power from bottom to top is the vibration area.

[0009] The number of crossings is defined as follows: starting from the jth feasible area of ​​the i-th unit, crossing the boundary from the feasible area to the vibration area once is counted as one crossing; crossing the boundary from the vibration area to the feasible area once is not counted as one crossing; crossing from the feasible area to the vibration area once and then crossing out from the other boundary of the vibration area with the feasible area is counted as one complete crossing; if the starting power and the final power are on both sides of the jth vibration area of ​​the i-th unit, it means that the vibration area has been completely crossed once in this power allocation. Figure 2 .

[0010] The definition of full span mileage is: only the mileage of the complete span of the vibration zone between the starting power and the final power is considered. Figure 3 .

[0011] The definition of the time passing through the vibration zone is: regardless of whether the unit directly crosses the vibration zone or the unit's final target power stays in the vibration zone and the next adjustment instruction is given before crossing out of the vibration zone, the vibration zone time is the time the unit's real-time power experiences from entering the vibration zone to finally crossing out of the vibration zone.

[0012] In step 3, four objective functions were set based on the objectives of the unit model: the priority of power allocation, the length of the complete span, the number of complete spans, and the number of units falling into the vibration zone. Equality and inequality constraints were set based on the power balance requirements of the entire AGC unit and the range and direction of power output of each unit.

[0013] (1) The power allocation priority objective function is:

[0014] After multiple rounds of clearing in the frequency modulation market, bids were submitted in descending order according to the frequency modulation mileage price, resulting in the winning AGC units for the operating period. The frequency modulation mileage price is based on the mileage quote reported by the units, taking into account the frequency modulation performance indicators obtained from each unit's actual regulation rate, response time, and regulation accuracy, and the final price is obtained. AGC units with low frequency modulation mileage prices are generally units with relatively good quotes and overall unit performance. When the actual AGC total power command is finally allocated, AGC units with high frequency modulation mileage prices are also given priority. The objective function is:

[0015]

[0016] Where F1 is the power allocation priority objective function; n is the number of AGC units participating in the AGC total power command allocation at the current command moment; Δx i =x i ′-x i Indicates the output change of the i-th unit, where x i and x i ′ respectively represent the initial power and final target power of the i-th unit; i is used to represent the unit ranking. Considering the priority, the lower the unit ranking, the higher the unit priority; u i Indicates whether unit i participates in the allocation of the AGC total power command. If yes, the value is 1; if not, the value is 0.

[0017] (2) The objective function of the complete mileage is:

[0018] When allocating the AGC total power command, if the total power that can be adjusted by each unit is less than the total power command without crossing the vibration zone, some units must be selected to cross the vibration zone to ensure that the total output meets the total power command. When this situation occurs where crossing the vibration zone is necessary, the objective function only considers the size of the complete crossing distance. It is hoped that the final allocation strategy will cross as little complete distance as possible. The objective function is:

[0019]

[0020] Where F2 is the objective function of the complete mileage span; n is the number of AGC units participating in the allocation of the AGC total power command at the current command moment; m-1 represents the total number of vibration zones of a winning AGC unit; P ij Indicates the value of the jth vibration zone of the i-th unit, refer to Figure 2 Schematic diagram of the unit operating area, xi(j+1)min and x ijmax They represent the upper and lower limits of the jth vibration zone of the i-th unit, respectively, and their expression is P ij =x i(j+1)min -x ijmax ;y ij Indicates whether the output of the i-th unit can completely pass through the j-th vibration zone after the output changes. At this time, the total power instruction ΔP must be considered. G direction.

[0021] when

[0022] when

[0023] If the final target power is in the vibration zone, it will be considered in the fourth objective function. This objective function only considers that the final objective function of each unit falls in the operating zone.

[0024] (3) The objective function of the number of complete spans is:

[0025] After determining that the result of the AGC total power command distribution will inevitably cause the final target power of some AGC units to cross the vibration zone, in addition to the second objective function's requirement for the complete crossing mileage, it is also hoped that the distribution result can also take into account the minimum number of complete crossings. The objective function is:

[0026]

[0027] Where F3 is the objective function of the number of complete crossings; the number of times each unit completely crosses the vibration zone is accumulated.

[0028] (4) The objective function of the number of units falling into the vibration zone is:

[0029] The objective functions (2) and (3) above both ignore the situation where the final target power falls within the vibration zone. If the final target power of the unit always falls within the vibration zone while satisfying the equality and inequality constraints, and it is also hoped that the number of units falling within the vibration zone is small, the objective function is:

[0030]

[0031] Where, F4 is the target function of the number of units falling in the vibration zone; ij Indicates whether the final target power of the i-th unit will fall into the j-th vibration zone, and its expression is Z ij =1 indicates the final target power x i ′ is located between the upper and lower limits of the jth vibration zone of the i-th unit, that is, the final target power falls within this vibration zone, Zij =0 means that the final target power does not fall within the vibration range.

[0032] However, if the starting power of some units falls within the vibration zone, even if z = 1, u = 0 will be set to minimize the objective function during the solution. As a result, the units that are stopped in the vibration zone will remain in this zone for a long time, which is unacceptable. Therefore, the model imposes a restriction on the input starting power: it cannot fall within the vibration zone. Power that falls within the vibration zone must first be corrected to exit the vibration zone before it can be input into the optimization model for solution.

[0033] The equality and inequality constraints are:

[0034] (1) Power balance equality constraint

[0035] The power balance requirement for all AGC units means that the sum of the power allocated to all winning AGC units is equal to the total power instruction ΔP G :

[0036]

[0037] (2) Output direction inequality constraint

[0038] In order to avoid the phenomenon of wasting power generation resources by increasing the output of some units and reducing the output of others, it is necessary to specify the direction of power distribution of all winning AGC units to be the same as the total power instruction ΔP G The same direction:

[0039] ΔP G Δx i u i ≥0

[0040] (3) Output numerical inequality constraints

[0041] The output value of each AGC unit is affected by the current state of the unit and the need to ensure system safety and stability. It should be noted that after the unit receives the adjustment command, the output changes, and the final target power will not fall outside the adjustment range of the winning bid:

[0042] x i1 min ≤x i +Δx i ≤x im max

[0043] Where x i1 min and x im max are the minimum and maximum outputs of the i-th unit during this operating period.

[0044] When power generation units declare frequency regulation capacity, in order to prevent the system power flow distribution from changing significantly due to frequency regulation, which may cause system instability, the sum of the frequency regulation capacity of each power generation unit in a single power plant cannot exceed 20% of the frequency regulation capacity demand value of the control area. Here, referring to the constraint rules for frequency regulation capacity declaration, when performing real-time AGC power allocation, to prevent the system stability from being affected by excessive power changes of some units, the output change Δx of the i-th unit is set i Does not exceed the total power command ΔP G 20%:

[0045] |Δx i |≤0.2*|ΔP G |

[0046] When the tie line deviation changes drastically, there will be two consecutive total power instructions ΔP G In the event of an adjustment in the opposite direction, if the previous instruction assigned to a unit was too large and the adjustment has not yet been completed, the adjustment direction of the unit will be opposite to the direction of the AGC unit that is currently hoping to win the bid. In this case, because the previous instruction was too large, not only will other units be idle, reducing the overall adjustment rate of the winning AGC unit, but when the second total power instruction is in the opposite direction, the unit that has not completed the adjustment last time will also play a role in "disrupting" the adjustment, ultimately wasting valuable adjustment resources. Therefore, another layer of restrictions is placed on the instructions assigned to the units to ensure that all units have completed the assigned adjustment tasks when the next instruction is generated:

[0047] |Δx i |≤|(Tt i )v i |

[0048] Where, T represents the interval between two AGC commands, which is 40s; t i and v i They represent the response time and regulation rate of the i-th unit respectively.

[0049] Combining the above equations, we can obtain an inequality constraint that limits the output change of the i-th unit:

[0050] |Δx i |≤min(0.2*|ΔP G |,|(Tt i )v i |)

[0051] Step 4: Analyze the priority of each objective function and set its coefficients based on their importance. This allows the multi-objective optimization problem to be transformed into a single-objective optimization problem using a weighted sum approach. Due to the power allocation strategy, the current initial power of each unit must be verified. If some units are within the vibration zone, the starting power is adjusted toward the total power to ensure that the initial power is outside the vibration zone. This ultimately results in the new unit initial power and total power.

[0052] like Figure 5 As shown in the figure, the priority applied to this multi-objective function is: power allocation priority < complete span mileage < complete span times < number of units falling into the vibration zone. The following three scenarios may be encountered during power allocation:

[0053] (1) When there is no need for some units to cross the vibration zone to meet the constraint of total power instruction ΔP G In the case of allocation, F2=F3=F4=0, the objective function is minF=aF1, and the power only needs to be allocated according to priority;

[0054] (2) When some units must cross the vibration zone, but the final target power does not need to stay in the vibration zone, F4 = 0. At this time, the objective function is min F = aF1 + bF2 + cF3. In the priority selection of the span mileage and the span number, our algorithm pays more attention to the span number. The corresponding coefficient c can be set very large, so that the optimization algorithm will give priority to finding a power allocation strategy that can reduce the span number, and then search for an allocation strategy with a small span mileage, and finally search for a strategy that allocates power according to priority.

[0055] (3) When some units must cross the vibration zone and a certain target power must eventually remain in the vibration zone, the objective function is min F = aF1 + bF2 + cF3 + dF4. At this time, we give priority to the allocation result that will minimize the number of units falling into the vibration zone. The corresponding coefficient d can be set to the largest of the four coefficients, and the remaining three coefficients are set as above (2).

[0056] Then determine the coefficient corresponding to each objective function.

[0057] Assume the minimum coefficient a=1.

[0058] In order to distinguish the priorities of F1 and F2 in scenario (2), there are:

[0059]

[0060] Therefore, Next, we distinguish the priorities of F3, F1, and F2.

[0061]

[0062] Therefore,

[0063] In order to distinguish the priority of F4 from F1, F2 and F3 in scenario (3), it is known that F4 has the highest priority, so d can be a large value. In the program, d is set to 10. 8 .

[0064] Therefore, the final single objective function minF=aF1+bF2+cF3+dF4, where d = 10 8 According to the equality and inequality constraints of the equation, the parameter information of the winning AGC unit, the total power instruction ΔP at each command time G The final distributed power result can be obtained.

[0065] Step 5: Solve the constructed model through the mathematical programming optimizer (Gurobi) to obtain the final allocation result.

[0066] Compared with the prior art, the advantages of the AGC total power allocation strategy considering the vibration zone in the frequency modulation market of the present invention are:

[0067] 1. Compared with the traditional power allocation strategy of allocating power according to the proportion of adjustable capacity and the power allocation strategy of prioritizing frequency modulation mileage from low to high, the allocation strategy of the present invention reduces the number of crossings and the time the unit stays in the vibration zone, thus reducing the loss of hydropower AGC units and the total cost of social operation.

[0068] 2. The allocation strategy of the present invention also fully mobilizes the enthusiasm of each unit with different frequency modulation mileage ranking prices. For different settings of the prohibited zone values, the allocation strategy can also solve the total power instruction allocation problem with multiple prohibited zones or no prohibited zones. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings in the following description only relate to some embodiments of the present invention and are not limitations of the present invention.

[0070] Figure 1 It is the overall flow chart.

[0071] Figure 2 This is a schematic diagram of the unit operating area.

[0072] Figure 3 It is a schematic diagram of various spanning relationships.

[0073] Figure 4 This is a complete mileage diagram.

[0074] Figure 5 is the priority relationship of each objective function.

[0075] Figure 6 It is the change of unit output under the control of strategy (1).

[0076] Figure 7 Is strategy (1) ΔP within 1 hour G Order the situation.

[0077] Figure 8 It is the change of unit output under the control of strategy (2).

[0078] Figure 9 It is strategy (2) ΔP within 1 hour G Order the situation.

[0079] Figure 10 It is the change of unit output under the control of strategy (3).

[0080] Figure 11 It is strategy (3) ΔP within 1 hour G Order the situation. DETAILED DESCRIPTION

[0081] In order to more clearly understand the above-mentioned objectives, features and advantages of the present invention, the technical solution of the present invention is further described in detail in a non-limiting manner in conjunction with the accompanying drawings and specific embodiments.

[0082] Based on the technical solution of the present invention, this implementation selected the Guangxi power grid for simulation analysis from 12:00 PM to 1:00 PM on December 28, 2018. Command values ​​were calculated and power allocation was performed every 40 seconds, for a total of 90 commands per hour. Information on the winning AGC units is shown in Table 1 below.

[0083] Table 1:

[0084]

[0085] The simulation analysis of Guangxi Power Grid was conducted from 12:00 to 13:00 on December 28, 2018. The command value was calculated and power distribution was performed every 40 seconds. A total of 90 commands were issued in one hour. Figure 5 The process performs calculations for AGC load frequency control and power generation distribution.

[0086] In order to verify the effectiveness of the AGC total power instruction allocation strategy considering the vibration zone, two comparative allocation strategies are set: Strategy (1), the traditional power proportion allocation according to the adjustable capacity; Strategy (2), the power allocation is performed only according to the priority of the frequency modulation mileage price from low to high without considering the vibration zone of each unit. This comparative strategy can be understood as a special case of the AGC total power instruction allocation strategy considering the vibration zone. Let the weight coefficient b = c = d = 0, then its objective function is simplified to: minF = F1, other constraints remain unchanged, but no power correction operation is required; Strategy (3), the allocation strategy proposed by the present invention. The results of the horizontal comparison of each allocation strategy are shown in Table 2 below.

[0087] Table 2:

[0088]

[0089]

[0090] As can be clearly seen from Table 2, Strategy (3) proposed in this invention uses the number of crossings and the time the unit remains in the vibration zone as objective functions, significantly reducing the number of crossings and the time the unit remains in the vibration zone. Strategy (1) and Strategy (2), on the other hand, do not impose any constraints on the vibration zone. The unit will arbitrarily cross or remain in the vibration zone, which has a significant impact on the safe and stable operation of the unit.

[0091] Strategy (1) for total power command ΔP G When allocating power, the unit performance and frequency regulation quotation are not considered, and the power allocation is based on the adjustable capacity only. Therefore, each time the total power instruction ΔP G When making adjustments, all the winning AGC units are mobilized for adjustment. The overall adjustment rate is very high, and the total power instruction ΔP can be completed as quickly as possible. G task; and strategy (2) for the total power instruction ΔP G When allocating, the total power instruction ΔP is sorted from low to high in order of frequency modulation mileage price. G Assigned to each winning AGC unit, each unit assigned to the regulation task bears the maximum regulation power under the current constraints, greatly reducing the number of units participating in each regulation and the overall regulation rate. Among the three allocation strategies, the slowest total power instruction ΔP GTask; Strategy (3) proposed in the present invention takes into account the existence of vibration zones based on Strategy (2). In order to reduce the number of times and mileage crossing the vibration zone, each unit assigned to the regulation task often does not have to bear the maximum regulation power under the current constraints. Compared with Strategy (2), the number of units participating in the regulation each time and the overall regulation rate increase. By compromising among the three allocation strategies, the overall regulation rate can be obtained: Strategy (1) > Strategy (3) > Strategy (2). The overall regulation rate is large, and it can quickly complete an adjustment task, control the power deviation as quickly as possible within the qualified range of the CPS indicator, and also indirectly reduce the possibility of continuing to issue commands later, reducing the number of instructions generated. Finally, the generated instruction ΔP is obtained by simulation statistics. G Number of times: Strategy (1) < Strategy (3) < Strategy (2).

[0092] As for changes in compensation costs, the calculation of unit compensation costs, as described in Formula (2-10), has no absolute relationship to the allocation strategy. This is because when determining the winning AGC units, the price ranked highest by frequency regulation mileage is used as the marginal price for the overall winning AGC unit. During settlement, the per-MW subsidy price for each unit is calculated by multiplying the marginal price by the respective unit's comprehensive frequency regulation performance index. Therefore, if a strategy allocates more units with lower comprehensive frequency regulation performance indexes, the final overall compensation cost will be relatively lower, but this is also related to the specific frequency regulation mileage allocated.

[0093] The control results of strategy (1) are shown in Table 3. Figure 6 and Figure 7 As shown in the figure, the dashed lines represent the upper and lower limits of each unit's vibration zone. The declared regulating capacity is closely related to the installed capacity of the generating unit. Units with larger installed capacity can claim a larger regulating capacity, and ultimately receive a larger regulating power. Furthermore, since each unit's regulating power is allocated based on its adjustable capacity, it receives subsidies commensurate with its installed capacity within the bidding cycle, preventing some winning units from receiving extremely low or even no compensation.

[0094] Table 3:

[0095]

[0096] The control results of strategy (2) are shown in Table 4. Figure 8 and Figure 9 Because the bids are sorted simply by the price of the frequency modulation mileage, it is obvious from the numerical value of the modulation mileage that power regulation is mainly undertaken by the units with low frequency modulation mileage prices, while the units with high frequency modulation mileage prices can only be allocated some modulation tasks when power compensation is urgently needed. As a result, the compensation amount allocated to the units with high frequency modulation mileage prices is far lower than that of other winning bidders.

[0097] Table 4:

[0098]

[0099] The control results of strategy (3) are shown in Table 5. Figure 10 and Figure 11 As shown. Compared with the previous two allocation strategies, Strategy (3) fully mobilizes the adjustment capabilities of each winning unit in order to reduce the number of crossings and the time the unit stays in the vibration zone. It also solves the problem in Strategy (1) that the compensation amount is too equal and only related to the size of the unit's installed capacity. It also solves the problem in Strategy (2) that the power allocation is too biased towards units with low frequency regulation mileage ranking prices. Finally, after allocation according to Strategy (3), the units generally only operate on one side of the vibration zone. Only when the power allocation cannot meet the demand will the units be arranged to cross to the other side of the vibration zone for adjustment. Since the vibration zone situation is taken into account based on the ranking price, the AGC units that won the bid but have a higher ranking price also have a greater chance of being allocated power, which fully mobilizes the enthusiasm of the units with different frequency regulation mileage ranking prices.

[0100] Table 5:

[0101]

[0102]

[0103] The present invention establishes a multi-objective optimization problem model for the allocation of AGC total power instructions taking into account the vibration zone in the frequency regulation market. By determining the importance of each objective function and analyzing the coefficients of each objective function, the multi-objective optimization problem can be smoothly converted into a single-objective optimization problem for solution. A "load frequency control + power generation distribution" AGC model was built, and simulation control was performed on an operating period. Compared with the traditional strategy of allocating power according to the adjustable capacity ratio and the power allocation strategy based only on the priority of the frequency regulation mileage ranking price from low to high, the strategy proposed by the present invention not only reduces the number of crossings and the time the unit stays in the vibration zone, reduces the loss of the hydropower AGC unit, and reduces the total cost of social operation, but also fully mobilizes the enthusiasm of each unit with different frequency regulation mileage ranking prices. For different settings of the prohibited zone values, the allocation strategy can also solve the total power instruction allocation problem with multiple prohibited zones or no prohibited zones.

Claims

1. An AGC total power allocation strategy considering vibration zones in a frequency modulation market, characterized by The following steps are included: Step 1: Based on the real-time data collected from the power grid and the overall regulation instruction strategy, the command value is obtained; Step 2: Build a unit model and divide the output operating range of the hydropower unit into different areas based on the difference between the feasible area and the vibration area. The areas are divided and numbered from bottom to top in terms of power to obtain the unit operating area. The concept of the number of spans, complete span mileage, and time spent in the vibration area involved in the constructed model is defined. Step 3: Based on the objectives of the unit model, four objective functions are set, namely the priority of power allocation, the size of the complete span mileage, the number of complete span times, and the number of units falling into the vibration zone; according to the power balance requirements of the entire AGC unit and the range and direction of the power output of each unit, equality and inequality constraints are set; among them, The mathematical expression of the objective function of the power allocation is: Where F1 is the power allocation priority objective function; n is the number of AGC units participating in the AGC total power instruction allocation at the current command moment; Δx i =x i ′-x i Indicates the output change of the i-th unit, where x i and x i 'represent the initial power and final target power of the i-th unit respectively; i is used to represent the unit ranking. Considering the priority, the lower the unit ranking, the higher the unit priority; u i Indicates whether unit i participates in the allocation of the AGC total power command. If yes, it is 1; if not, it is 0. The mathematical expression of the objective function of the complete mileage is: Where F2 is the objective function for the complete mileage span; n is the number of AGC units participating in the AGC total power command allocation at the current command moment; m-1 represents the total number of vibration zones of a winning AGC unit; u i Indicates whether unit i participates in the allocation of the AGC total power command. If yes, it is 1; if no, it is 0. ij Indicates the value of the jth vibration zone of the i-th unit, x i(j+1)min and x ijmax They represent the upper and lower limits of the jth vibration zone of the i-th unit, respectively. The expression is P ij =x i(j+1)min -x ijmax ;y ij Indicates whether the i-th unit can completely pass through the j-th vibration zone after the output changes; The mathematical expression of the objective function of the number of complete spans is: Where F3 is the objective function of the number of complete crossings, and the number of times each unit completely crosses the vibration zone is accumulated; The mathematical expression of the objective function of the number of units falling in the vibration zone is: Where, F4 is the target function of the number of units falling in the vibration zone; ij Indicates whether the final target power of the i-th unit will fall within the j-th vibration zone; Step 4: Analyze the priorities of the four objective functions, set the coefficients of the objective functions according to their importance, and transform the multi-objective optimization problem into a single-objective optimization problem using the weighted sum method for solution. Verify the current initial power of each unit. If some units are in the vibration zone, adjust the starting power towards the total power so that the initial power is outside the vibration zone. Finally, obtain the new unit initial power and total power. Step 5: Solve the constructed model through the mathematical programming optimizer to obtain the final allocation result.

2. The AGC total power allocation strategy considering vibration zones in the frequency modulation market according to claim 1 is characterized in that: In the step 3, y ij Indicates whether the output of the i-th unit can completely pass through the j-th vibration zone after the output changes, and the total power command ΔP G Directions related to: When ΔP G >0, When ΔP G <0, If the final target power is in the vibration zone, it will be considered in the fourth objective function. This objective function only considers that the final objective function of each unit falls in the operating zone; z ij Indicates whether the final target power of the i-th unit will fall into the j-th vibration zone. Its expression is: z ij =1 indicates the final target power x i ' is between the upper and lower limits of the jth vibration zone of the i-th unit, that is, the final target power falls within this vibration zone, z ij =0 means that the final target power does not fall within the vibration range.

3. The AGC total power allocation strategy considering vibration zones in the frequency modulation market according to claim 2, characterized in that: In step 3, the equality and inequality constraints include: Power balance equality constraints: The sum of the power allocated to all winning AGC units is equal to the total power command ΔP G , the expression is: Output direction inequality constraint: The power distribution direction of all winning AGC units is the same as the total power instruction ΔP G The direction is the same, and the expression is: ΔP G Δx i u i ≥0 Output value inequality constraints: After receiving the adjustment command, the unit changes its output, and the final target power will not fall outside the adjustment range of the winning bid. The expression is x i1 min ≤x i +Δx i ≤x im max Where x i1 min and x im max are the minimum and maximum outputs of the i-th unit during this operating period.

Images