Optimization method for multi-time-scale energy-saving operation of urban rail power supply system

Abstract

The invention discloses an optimization method for multi-time-scale energy-saving operation of an urban rail power supply system, relates to the field of rail transit power supply, and solves the technical problems that response is delayed due to the fact that an existing urban rail power supply system adopts main transformer on-load voltage regulation and static var generator actions, equipment aging is easily accelerated due to frequent actions of tapping points, and the SVG capacity is limited. The optimization method comprises the following steps: acquiring an operation optimization stage of the urban rail power supply system, and when the operation optimization stage of the urban rail power supply system is a day-ahead operation optimization stage, triggering a day-ahead optimization strategy to generate a day-ahead optimization result; and when the operation optimization stage of the urban rail power supply system is an intra-day operation optimization stage, triggering an intra-day optimization strategy according to a day-ahead optimization result to obtain the total traction energy consumption of the power supply system and the highest steel rail potential of the target whole line. The problems of network voltage fluctuation and existence of a large amount of reactive power in an urban rail power supply system are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail transit power supply, more particularly, to a kind of urban rail transit power supply system multi-time scale energy-saving operation optimization method. BACKGROUND

[0002] With the acceleration of urbanization process, urban rail transit has become the backbone of modern urban public transport system due to its large capacity, high efficiency and low carbon environmental protection. However, as the core energy support for train operation, urban rail transit power supply system is facing increasingly severe technical challenges. On the one hand, train operation load presents strong time-varying and random characteristics, such as frequent start-stop and instantaneous feedback of regenerative braking energy, which leads to intensified traction network voltage fluctuation and dynamic changes in reactive power demand. On the other hand, energy shortage and the "double carbon" target have put forward higher requirements for energy saving and consumption reduction of the power supply system.

[0003] Currently, urban rail transit power supply system generally uses main transformer on-load voltage regulation and static var generator to deal with traction network voltage fluctuation and the problem of large amount of reactive power in the system. The main transformer maintains the bus voltage stable through tap adjustment, but its action has response delay, which is difficult to cope with dynamic voltage fluctuation. Meanwhile, frequent action of tap can accelerate equipment aging and increase operation and maintenance cost. Although SVG has the ability of dynamic reactive power compensation in seconds, its capacity is limited, and it is easy to have regulation capacity saturation or over-compensation problem in extreme load fluctuation scenarios. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a kind of urban rail transit power supply system multi-time scale energy-saving operation optimization method to solve the technical problems of response delay of main transformer on-load voltage regulation and static var generator action, easy acceleration of equipment aging by frequent action of tap and limited capacity of SVG in the existing urban rail transit power supply system.

[0005] The urban rail transit power supply system multi-time scale energy-saving operation optimization method comprises: obtaining an urban rail transit power supply system operation optimization stage; when the urban rail transit power supply system operation optimization stage is a day-ahead operation optimization stage, triggering a day-ahead optimization strategy to generate a day-ahead optimization result; when the urban rail transit power supply system operation optimization stage is an intra-day operation optimization stage, triggering an intra-day optimization strategy according to the day-ahead optimization result to obtain total traction energy consumption of the power supply system and target maximum rail potential of the whole line.

[0006] The day-ahead optimization strategy comprises:

[0007] S11: load the i-th operation period and the departure interval D ian operation diagram, inputting required basic data including line parameters, locomotive characteristic parameters and power supply system information according to the operation diagram, setting a main transformer on-load voltage regulation ratio K i , setting a constraint condition according to the basic data

[0008] S12: setting a population number N, an iteration number T, a traction energy consumption weight ω1 and a rail potential weight ω2, and generating a day-ahead optimization model by using a salp swarm algorithm and according to the main transformer on-load voltage regulation ratio K i , the population number N, the iteration number T, the traction energy consumption weight ω1 and the rail potential weight ω2

[0009] S13: outputting a reactive compensation value Q i of an SVG by the day-ahead optimization model i , constructing an objective function F(K i ,Q i ) according to the reactive compensation value Q i of the SVG and the main transformer on-load voltage regulation ratio K i , converting the objective function F(K i ,Q i ) into a single objective function F'(K i ,Q i ) by using linear normalization and fixed weight methods

[0010] S14: setting an iteration convergence condition, counting maximum and minimum values of traction energy consumption and rail potential of all individuals in the population, performing fitness sorting on the single objective function F'(K i ,Q i ) and judging whether the single objective function F'(K i ,Q i ) after sorting meets the iteration convergence condition, and returning to S12 to reset the main transformer on-load voltage regulation ratio K i when there is the single objective function F'(K i ,Q i ) that does not meet the iteration convergence condition

[0011] S15: when the iteration convergence condition is met, comparing objective function values F i under all the main transformer on-load voltage regulation ratios K i , outputting an optimal solution and a reactive compensation value Q day-ahead by the target day-ahead optimization model, and taking the optimal solution and the reactive compensation value Q day-ahead as a day-ahead optimization result

[0012] the day-ahead optimization strategy includes

[0013] S21: Load the short-circuit capacity, grid connection impedance, and load demand within the prediction window p from the day-ahead optimization results; obtain train load prediction data within a short time scale; optimize the period T based on the short-circuit capacity, grid connection impedance, and load demand within the prediction window p, using the train load prediction data within the short time scale. s The length of the prediction window p;

[0014] S22: Identify the external power supply parameters, and based on the external power supply parameters, short-circuit capacity, grid impedance, and load demand within the prediction window, obtain the values ​​from time τ to time τ+T within the prediction window p. s External power supply voltage U at time s (τ+T s |τ);

[0015] S23: According to the external power supply voltage U s (τ+T s |τ) and combined with the prediction window p from time τ to time τ+T s The train load demand at any given time is predicted, and an intraday optimization model is obtained using the Zunhaishao algorithm. The intraday optimization model outputs the reactive power compensation prediction value Q of SVG. i intry-dad (τ+T s |τ);

[0016] S24: Select the reactive power compensation prediction value Q of the SVG. i intry-dad (τ+T s |τ) is used as the SVG output task within the prediction window and then assigned to the SVG to update the reactive power compensation value of each device; τ+T within the prediction window p is used as the SVG output task. s Time and departure interval D i The comparison is made when τ+T is within the prediction window p. s The time interval is less than or equal to the departure interval D. i If the condition is met, return to S21 to reload the short-circuit capacity, grid impedance, and load demand within the prediction window p from the previous day's optimization results; otherwise, calculate the total traction energy consumption of the power supply system and the target maximum rail potential for the entire line during the i-th operating period.

[0017] As a further improvement, in S11, the objective function F(K) i Q i The expression for ) is:

[0018] F(K i Q i )=min[f1(K i Q i ),f2(K i Q i]);

[0019] wherein f1(K i ,Q i ) is the total traction energy consumption of the power supply system in the i-th operation period, f2(K i ,Q i ) is the maximum rail potential in the i-th operation period; K i is the set of on-load tap ratios of all main transformers in the i-th operation period, K i ={K i,j |j=1,2,…,N MT}, N MT is the number of main transformers in the power supply system information; Q i is the set of SVG reactive compensation values in the i-th operation period,

[0020] Q i ={Q i,j |j=1,2,…,N SVG}, N SVG is the number of SVGs in the power supply system.

[0021] Further, the expression for calculating the total traction energy consumption of the power supply system in the i-th operation period is:

[0022]

[0023] wherein W T (t,K i,j ,Q i,j ) is the total active energy of the traction substation in the rectification mode at the j-th main transformer in the t-th second of the i-th operation period, W F (t,K i,j ,Q i,j ) is the total active energy fed back by the inversion feedback device at the j-th main transformer in the t-th second of the i-th operation period, and W R (t,K i,j ,Q i,j ) is the amount of returned electricity at the j-th main transformer in the t-th second of the i-th operation period.

[0024] Further, the expression for calculating the maximum rail potential of the entire line in the i-th operation period is,

[0025]

[0026] wherein U ts,m (t,K i ,Q i ) is the maximum rail potential of the j-th main transformer in the t-th second of the i-th operation period.) is the rail potential at the mth traction substation node at the tth second of the ith operation period, m = 1, 2, …, N ts , N ts is the number of traction substations, U tr,n (t, K i , Q i ) is the rail potential at the n th train node at the tth second of the ith operation period, n = 1, 2, …, N tr , N tr is the number of up and down trains in the line parameter at the moment.

[0027] Further, in the S13, the expression of the single objective function F'(K i , Q i ) is,

[0028]

[0029] Wherein, ω1, ω2 are the traction energy consumption weight of the power supply system and the weight of the highest rail potential of the whole line respectively, f 1max (Ki, Qi) is the maximum value of the traction energy consumption of the power supply system in each iteration of all individuals in the ith operation period, f 1min (Ki, Qi) is the minimum value of the traction energy consumption of the power supply system in each iteration of all individuals in the ith operation period, f 2max (Ki, Qi) is the maximum value of the highest rail potential of the whole line in each iteration of all individuals in the ith operation period, f 2min (Ki, Qi) is the minimum value of the highest rail potential of the whole line in each iteration of all individuals in the ith operation period.

[0030] Further, in the S12, the constraint conditions of the day-ahead stage include traction network voltage constraint condition, on-load tap changer position constraint condition, on-load tap changer total adjustment times constraint condition and SVG adjustable reactive power compensation value constraint condition.

[0031] Further, the expression of the traction network voltage constraint condition is:

[0032]

[0033] Wherein, U TN，x (t) is the traction network voltage at any line position x at the tth second in the power supply system information under normal operation, is the upper limit value of the traction network voltage, is the lower limit value of the traction network voltage.

[0034] Further, the expression of the on-load regulating transformer tap position constraint condition is

[0035]

[0036] Wherein, N T,,j is the number of tap positions of the jth main transformer on-load regulating transformer, H j is the maximum adjustable tap position number of the jth main transformer on-load regulating transformer in the adjacent operation period.

[0037] Further, the total regulation times of the on-load regulating transformer tap position throughout the day constraint condition is

[0038]

[0039] Wherein, S i,j is the on-load regulating transformer tap position regulation state value of the jth main transformer in the ith operation period, M i,j is the maximum allowable regulation times of the jth main transformer on-load regulating transformer tap position in the ith operation period, N D is the total operation period number of the line parameter throughout the day.

[0040] Further, the expression of the SVG adjustable reactive power compensation value constraint condition is

[0041]

[0042] Wherein, Q i,j is the adjustable reactive power compensation value of the jth SVG in the ith operation period, is the upper limit of the adjustable reactive power compensation value of the jth SVG in the ith operation period, is the lower limit of the adjustable reactive power compensation value of the jth SVG in the ith operation period.

[0043] Beneficial effects

[0044] The advantages of the present application are:

[0045] This invention obtains the operation optimization stage of the urban rail power supply system. When the operation optimization stage is the day-ahead operation optimization stage, a day-ahead optimization strategy is triggered to generate the day-ahead optimization results. When the operation optimization stage is the intraday operation optimization stage, an intraday optimization strategy is triggered based on the day-ahead optimization results to obtain the total traction energy consumption of the power supply system and the target maximum rail potential along the entire line. The invention addresses the problems of grid voltage fluctuations and a large amount of reactive power in the urban rail power supply system through a collaborative control architecture of on-load tap changer and SVG reactive power compensation. At the same time, it is necessary to consider the different action response times of the on-load tap changer and SVG device, as well as the constraints of tap position and the number of adjustments throughout the day, to reduce the system's traction energy consumption and rail potential, providing technical support for the green and low-carbon operation of urban rail transit. Attached Figure Description

[0046] Figure 1 This is a flowchart of the optimization method for multi-timescale energy-saving operation of the urban rail power supply system according to the present invention;

[0047] Figure 2 This is a flowchart of the sea tunic group algorithm of the present invention;

[0048] Figure 3 This is a convergence characteristic diagram of the optimization algorithm in an embodiment of the present invention;

[0049] Figure 4 This is the peak-period SVG output curve in an embodiment of the present invention;

[0050] Figure 5 This is an example of the on-load tap change curve of the main transformer in an embodiment of the present invention;

[0051] Figure 6 This is a diagram showing the location information of substations along an actual subway line in an embodiment of the present invention.

[0052] Figure 7 This is a diagram showing train information of an actual subway line in an embodiment of the present invention;

[0053] Figure 8 This is a diagram showing the main transformer information of an actual subway line in an embodiment of the present invention;

[0054] Figure 9 This is a schematic diagram illustrating the day-ahead optimization results of the on-load tap changer position of the main transformer and the reactive power compensation value of the SVG in an embodiment of the present invention;

[0055] Figure 10 This is a schematic diagram showing the total energy consumption of the all-day power supply system and the highest rail potential in an embodiment of the present invention. Detailed Implementation

[0056] The present invention will be further described below with reference to embodiments, but this does not constitute any limitation on the present invention. Any limited modifications made by any person within the scope of the claims of the present invention are still within the scope of the claims of the present invention.

[0057] See Figures 1-10 This invention discloses an optimization method for multi-timescale energy-saving operation of an urban rail power supply system, comprising a day-ahead operation optimization stage and an intraday operation optimization stage. The specific steps are as follows:

[0058] Step 1: Input the necessary basic data, such as line parameters, locomotive characteristic parameters, and power supply system information.

[0059] Step 2: Construct an optimization model for the energy-saving operation of the urban rail power supply system in the current phase, including the optimization variable: the on-load tap changing ratio K of the main transformer. i The reactive power compensation value Q of SVG i Objective function: The total traction energy consumption of the power supply system and the highest rail potential along the entire line are the optimization objectives F(K). i Q i Constraints: Traction network voltage U TN On-load tap changer tap position T i,j And the maximum number of adjustments per day M j Constraints, SVG reactive power compensation value Q j Constraints, etc.

[0060] Step 3: Transform the multi-objective function into a single-objective function F'(K) by using linear normalization, fixed weights, etc. i Q i Set the on-load tap changing ratio K of the main transformer. i The population size N, iteration number T, traction energy consumption weight ω1, and rail potential weight ω2 are set, and the daytime optimization model is solved using the tunic swarm algorithm.

[0061] Step 4: Calculate the maximum and minimum values ​​of medium energy consumption and rail potential for all individuals, calculate the normalized objective function value, sort by fitness, and determine whether the convergence condition is met.

[0062] Step 5: Compare the on-load tap changing ratio K of all main transformers i The objective function value F under the following conditions i The optimal solution and reactive power compensation value Q for the current stage are obtained. day-ahead .

[0063] Step 6: Control during the intraday operation optimization phase is based on the day-ahead optimization results, utilizing train load forecast data within a short timescale, and optimizing the cycle T within a short timescale. s To scroll the window forward.

[0064] Step 7: The control objective for intraday operation optimization is the total traction energy consumption F(Q) of the power supply system within the forecast window. i (τ+T s The constraints are similar to those in the daytime stage, and the optimization variable is the reactive power compensation value Q of the SVG. i .

[0065] Step 8: Combine time τ to time τ+T s Train load forecast demand at any given time, so that the power supply system can predict demand within the rolling time window T. s Taking the total traction energy consumption within the day as the objective, an improved tunic swarm algorithm based on parallel computing is used to solve the intraday optimization model, and the reactive power compensation prediction value Q of SVG is obtained. i intry-dad (τ+T s |τ).

[0066] Objective function for the current stage:

[0067] F(K i Q i )=min[f1(K i Q i ),f2(K i Q i )];

[0068] In the formula, f1(K) i Q i f2(K) i Q i ) represent the total traction energy consumption of the power supply system and the highest rail potential during the i-th operating period, respectively. i Let K be the set of on-load tap-changing ratios of all main transformers during the i-th operating period. i ={K i,j |j=1,2,…,N MT}, N MT Q represents the number of main transformers in the power supply system. i Let Q be the set of all SVG reactive power compensation values ​​during the i-th operating period. i ={Q i,j |j=1,2,…,N SVG}, N SVG This represents the number of SVGs in the power supply system.

[0069] The traction energy consumption of the power supply system in the i-th operating period is shown in the following formula:

[0070]

[0071] Where, f1(K) i Q iW represents the traction energy consumption of the power supply system during the i-th operating period. T (t,K i,j Q i,j W represents the total active power of the traction substation in the zone where the j-th main transformer is located during the t-th second of the i-th operating period, under rectification conditions. F (t,K i,j Q i,j W represents the total active power fed back by the inverter feedback device in the zone where the j-th main transformer is located during the i-th operating period at second t. R (t,K i,j Q i,j ) represents the returned power at the j-th main transformer in the t-th second of the i-th operating period.

[0072] The highest rail potential of the power supply system across the entire line during the i-th operating period is shown in the following formula:

[0073] f2(K i Q i ) = max(U ts，m (t,K i Q i )|,|U tr，n (t,K i Q i )|);

[0074] In the formula, f2(K) i Q i U is the highest rail potential of the power supply system along the entire line during the i-th operating period. ts,m (t,K i Q i Let be the rail potential at the m-th traction substation node in the t-th second of the i-th operating period, where m = 1, 2, ..., N. ts N ts U represents the number of traction substations. tr,n (t,K i Q i Let be the rail potential at the nth train node in the tth second of the i-th operating period, where n = 1, 2, ..., N. tr N tr This refers to the number of up and down trains in the line parameters at that moment.

[0075] Constraints for the current phase:

[0076] (1) Traction network voltage constraint

[0077]

[0078] Among them, U TN,x(t) represents the traction network voltage at any line position x in the power supply system information at second t during normal operation. This is the upper limit of the traction network voltage. This is the lower limit of the traction network voltage.

[0079] (2) Tap position constraints of on-load tap-changing transformers

[0080]

[0081] Where, N T,,j H represents the number of tap positions of the on-load tap changer of the j-th main transformer. j Let be the maximum number of adjustable taps of the on-load tap changer of the j-th main transformer during adjacent operating periods.

[0082] (3) Constraint on the total number of tap changes per day for on-load tap-changing transformers

[0083]

[0084] Among them, S i,j M represents the on-load tap changer adjustment status value of the j-th main transformer during the i-th operating period in the power supply system information. i,j N represents the maximum permissible number of daily adjustments made by the on-load tap changer of the j-th main transformer in the power supply system information. D This refers to the total number of operating hours throughout the day in the route parameters.

[0085] (4) Adjustable reactive power compensation value constraint of SVG

[0086]

[0087] Among them, Q i,j Let j be the adjustable reactive power compensation value for the j-th SVG. This represents the upper limit of the reactive power compensation value for the j-th SVG. This is the lower limit of the reactive power compensation value for the j-th SVG.

[0088] Normalized objective function value:

[0089]

[0090] Wherein, ω1 and ω2 are the weights of the traction energy consumption of the power supply system and the highest rail potential along the entire line, respectively, and f 1max (Ki,Qi) Maximum value, f 1min (Ki, Qi) represents the minimum traction energy consumption of the power supply system in each iteration during the i-th operating period, f. 2max (Ki, Qi) represents the maximum value of the highest rail potential across the entire line in each iteration during the i-th operating period, and f2min (Ki,Qi) represents the minimum value of the highest rail potential across the entire line during the i-th operating period, among all individuals in each iteration.

[0091] The objective function and constraints for the intraday phase are similar to those for the day-ahead phase, and will not be elaborated further.

[0092] To better understand the optimization method for multi-timescale energy-saving operation of urban rail power supply systems described above, the present invention provides the following specific embodiments:

[0093] Taking a certain actual subway line as an example, the line is 24km long and has 2 main substations, 10 traction substations, and 10 step-down substations. There are stations at each substation, and the trains are of type "6B".

[0094] The actual location information of the substations along the subway line is as follows: Figure 6 As shown. Train information is as follows. Figure 7 As shown, Figures 4-5 As shown, the main transformers MT in main substations MS1 and MS2 are both on-load tap-changing transformers, numbered MT1 to MT4 respectively. Furthermore, the low-voltage busbars of these main transformers are equipped with reactive power compensation devices (SVG), numbered SVG1 to SVG4 respectively. The main substation information is as follows: Figure 7 As shown.

[0095] like Figure 3 As shown, during the current operation optimization phase, the weights for traction energy consumption of the power supply system and the highest rail potential along the entire line were set to 0.8 and 0.2, respectively. The population size was set to 50, and the number of iterations was set to 100. The optimization solution was performed during peak hours when the train interval was 240 seconds. The iteration process is as follows: Figure 2 As shown, when the number of iterations reaches 40, the normalized objective function value basically converges to about 0.01.

[0096] The optimization results of the on-load tap changer positions of the main transformer and the reactive power compensation value of the SVG are as follows: Figure 8 As shown.

[0097] During the intraday operation optimization phase, the set of on-load tap-changing ratios of each main transformer obtained in the previous day's operation optimization phase is used as the on-load tap-changing ratio of each main transformer in the intraday operation optimization phase. The reactive power compensation values ​​of each SVG obtained in the previous day's operation optimization phase are used as the initial values ​​for SVG optimization in the intraday operation optimization phase. The highest rail potential across the entire line obtained in the previous day's operation optimization phase is used as the highest rail potential constraint in the intraday operation optimization phase. A short-time scale period T is set. S The peak SVG output curve is obtained as follows: (The curve is 5 seconds long). Figure 9 As shown.

[0098] The maximum number of times the main transformer can be adjusted throughout the day is set to 10. Optimization solutions are performed for different departure intervals within the daily train schedule to obtain the on-load tap changer positions of each main transformer during the entire day's operating hours, as follows: Figure 10 As shown. The system before optimization is defined as the reference system, the system that only performs daytime operation optimization control is defined as Case 1, and the system that performs daytime and intraday multi-timescale operation optimization control is defined as Case 2. The total energy consumption and peak rail potential of the power supply system during the entire day's operation are statistically obtained according to the daily train schedule, as shown below. Figure 10 As shown.

[0099] The total number of on-load tap changers for main transformers MT1 to MT4 during the entire day's operation was 4, 4, 5, and 6, respectively, all of which met the daily limit for the total number of adjustments set by the constraints.

[0100] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention, and these will not affect the effectiveness of the implementation of the present invention or the practicality of the patent.

Claims

1. An optimization method for multi-timescale energy-saving operation of urban rail power supply systems, characterized in that, The optimization method includes obtaining the operation optimization stage of the urban rail power supply system, and when the operation optimization stage of the urban rail power supply system is the day-ahead operation optimization stage, triggering the day-ahead optimization strategy to generate the day-ahead optimization result. When the urban rail power supply system operation optimization phase is the intraday operation optimization phase, the intraday optimization strategy is triggered based on the previous day's optimization results to obtain the total traction energy consumption of the power supply system and the target highest rail potential along the entire line.

2. The optimization method for multi-timescale energy-saving operation of urban rail power supply system according to claim 1, characterized in that, The aforementioned optimization strategies include, S11: Load the i-th operating period and the departure interval as D. i Based on the operating diagram, input the required basic data and set the on-load tap changing ratio K of the main transformer. i Set constraints based on the aforementioned basic data; S12: Set the population size N, iteration number T, traction energy consumption weight ω1, and rail potential weight ω2, and use the Zunhaiqiao group algorithm based on the on-load tap-changing ratio K of the main transformer. i The day-ahead optimization model is generated by defining the constraints, population size N, iteration count T, traction energy consumption weight ω1, and rail potential weight ω2. S13: The reactive power compensation value Q of the SVG output by the day-ahead optimization model. i According to the reactive power compensation value Q of the SVG i and the on-load tap changing ratio K of the main transformer i Construct the objective function F(K) i Q i ), the objective function F(K) i Q i Transform into a single objective function F'(K) i Q i ); S14: Set the iterative convergence condition, and statistically analyze the maximum and minimum values ​​of traction energy consumption and rail potential for all individuals in the population. Apply the single objective function F'(K) i Q i Perform fitness ranking and determine the single objective function F'(K) after ranking. i Q i Does it satisfy the iterative convergence condition? If so, the single objective function F'(K) exists. i Q i If the iterative convergence condition is not met, then return to S12 to reset the on-load tap changing ratio K of the main transformer. i ; S15: When the iterative convergence condition is met, compare the on-load tap-changing ratio K of all main transformers. i The objective function value F under the following conditions i The target day-ahead optimization model outputs the optimal solution and the reactive power compensation value Q. day-ahead The optimal solution and the reactive power compensation value Q are combined. day-ahead This is the result of the recent optimization.

3. The optimization method for time-scale energy-saving operation of urban rail power supply system according to claim 2, characterized in that, In S11, the basic data includes line parameters, locomotive characteristic parameters, and power supply system information.

4. The optimization method for multi-timescale energy-saving operation of urban rail power supply system according to claim 2, characterized in that, In S11, the constraints include traction network voltage constraints, on-load tap changer tap position constraints, total number of tap changes per day constraints, and SVG adjustable reactive power compensation value constraints.

5. The optimization method for multi-time-scale energy-saving operation of urban rail power supply system according to claim 2, characterized in that, In S13, linear normalization and fixed weighting are used to transform the objective function F(K) i Q i This is transformed into a single objective function F'(Ki,Qi).

6. The optimization method for multi-timescale energy-saving operation of urban rail power supply system according to claim 2, characterized in that, In S13, the objective function F(K) is constructed. i Q i The expression for ) is: F(K i ,Q i )=min[f1(K i ,Q i ),f2(K i ,Q i )]; In the formula, f1(K) i Q i f2(K) represents the total traction energy consumption of the power supply system during the i-th operating period. i Q i K represents the highest rail potential during the i-th operating period; i Let K be the set of on-load tap-changing ratios of all main transformers during the i-th operating period. i ={K i,j |j=1,2,…,N MT }, N MT Q represents the number of main transformers in the aforementioned basic data; i Let Q be the set of all SVG reactive power compensation values ​​during the i-th operating period. i ={Q i,j |j=1,2,…,N SVG }, N SVG The number of SVGs in the power supply system.

7. The optimization method for multi-time-scale energy-saving operation of urban rail power supply system according to claim 2, characterized in that, In S13, the single objective function F'(K) i Q i The expression for ) is, Where ω1 is the traction energy consumption weight of the power supply system, ω2 is the weight of the highest rail potential along the entire line, and f 1max (Ki, Qi) represents the maximum value of the traction energy consumption of the power supply system in each iteration during the i-th operating period, and f is the maximum value of the traction energy consumption of the power supply system in each iteration. 1min (Ki,Qi) represents the minimum traction energy consumption of the power supply system during the i-th operating period in each iteration, f 2max (Ki, Qi) represents the maximum value of the highest rail potential across the entire line in each iteration during the i-th operating period, and f 2min (Ki,Qi) represents the minimum value of the highest rail potential across the entire line during the i-th operating period, among all individuals in each iteration.

8. The optimization method for multi-timescale energy-saving operation of urban rail power supply system according to claim 2, characterized in that, The intraday optimization strategy includes, S21: Load the short-circuit capacity, grid connection impedance, and load demand within the prediction window p from the day-ahead optimization results; obtain train load prediction data within a short time scale; optimize the period T based on the short-circuit capacity, grid connection impedance, and load demand within the prediction window p, using the train load prediction data within the short time scale. s The length of the prediction window p; S22: Identify the external power supply parameters, and based on the external power supply parameters, short-circuit capacity, grid impedance, and load demand within the prediction window, obtain the values ​​from time τ to time τ+T within the prediction window p. s External power supply voltage U at time s (τ+T s |τ); S23: According to the external power supply voltage U s (τ+T s |τ) and combined with the prediction window p from time τ to time τ+T s The train load demand at any given time is predicted, and an intraday optimization model is obtained using the Zunhaishao algorithm. The intraday optimization model outputs the reactive power compensation prediction value Q of SVG. i intry-dad (τ+T s |τ); S24: Select the reactive power compensation prediction value Q of the SVG. i intry-dad (τ+T s |τ) is used as the SVG output task within the prediction window and then assigned to the SVG to update the reactive power compensation value of each device; τ+T within the prediction window p is used as the SVG output task. s Time and departure interval D i The comparison is made when τ+T is within the prediction window p. s The time interval is less than or equal to the departure interval D. i If the condition is met, return to S21 to reload the short-circuit capacity, grid impedance, and load demand within the prediction window p from the previous day's optimization results; otherwise, calculate the total traction energy consumption of the power supply system and the target maximum rail potential for the entire line during the i-th operating period.

9. The optimization method for time-scale energy-saving operation of urban rail power supply system according to claim 8, characterized in that, The expression for calculating the total traction energy consumption of the power supply system during the i-th operating period is as follows: Among them, W T (t,K i,j Q i,j W represents the total active power of the traction substation in the zone where the j-th main transformer is located during the t-th second of the i-th operating period is in rectification mode. F (t,K i,j Q i,j W represents the total active power fed back by the inverter feedback device in the zone where the j-th main transformer is located during the i-th operating period at second t. R (t,K i,j Q i,j ) represents the returned power at the j-th main transformer in the t-th second of the i-th operating period.

10. The optimization method for time-scale energy-saving operation of an urban rail power supply system according to claim 9, characterized in that, The expression for calculating the highest rail potential across the entire line during the i-th operating period is as follows: f2(K i ,Q i )=max(|U t,m (t,K i ,Q i )|,|U tr，n (t,K i ,Q i )|); Among them, U ts,m (t,K i Q i Let be the rail potential at the m-th traction substation node in the t-th second of the i-th operating period, where m = 1, 2, ..., N. ts N ts U represents the number of traction substations. tr,n (t,K i Q i Let be the rail potential at the nth train node in the tth second of the i-th operating period, where n = 1, 2, ..., N. tr N tr This refers to the number of up and down trains in the line parameters at that moment.

Images