A train operation energy-saving method based on dynamic programming and time allocation
Through dynamic planning and time allocation, the problem of inaccurate model analysis when combining ATO speed curve and operation chart in the existing technology is solved, and the energy-saving effect of reducing train traction energy consumption and on-time operation is achieved, achieving 10.15% energy-saving effect .
Patent Information
- Application Number
- CN202010715808.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-07-23
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2040-07-23
AI Technical Summary
When the existing technology combines the train ATO speed curve and operating chart for integrated energy saving, the model analysis is inaccurate and fails to achieve true comprehensive energy saving effect.
Using a method based on dynamic programming and time allocation, a multi-stage calculation model for train traction energy consumption is established, and the interval operation curve of fixed time is searched through dichotomy, the time energy consumption inverse proportional characteristic curve is fitted, the operation time of each interval is optimized, and the redundant time is allocated. To reduce energy consumption, prepare new energy-saving operation diagrams.
The train operation curve and diagram are optimized and adjusted, ensuring on-time operation, reducing train traction energy consumption, and achieving 10.15% energy-saving effect on the software simulation platform.
Smart Images

Figure CN113971472B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of urban rail transit traction energy saving, in particular to a train operation energy saving method based on dynamic programming and time allocation. Background Art
[0002] With the rapid economic development and the vigorous promotion of urbanization, the scale of cities in my country has continued to expand, and the urban population and passenger volume have increased year by year. Conventional public transportation has been unable to meet the actual needs of my country's development. At this time, urban rail transit has stood out from a variety of public transportation tools with its large capacity, fast speed, less pollution, high punctuality, good comfort, and less road space per capita. It has not only improved transportation efficiency, but also greatly alleviated the congestion problem caused by the rapid increase in urban population. Therefore, more and more cities have begun to vigorously develop and build urban rail transit.
[0003] Compared with other types of urban public transportation, under the same transportation capacity conditions, the energy consumption of urban rail transit is about half of that of buses and one-ninth of that of cars. Obviously, the energy consumption of urban rail transit is relatively small compared with other modes of transportation, but considering the current large-scale construction and development, its total operating energy consumption will also be very large. Taking Beijing and Shanghai as examples, according to statistics, in 2017, the total electricity consumption of Beijing and Shanghai urban rail transit was about 1.78 billion kWh. In 2018, the total electricity consumption of Beijing's urban rail transit was about 1.9 billion kWh, and the total electricity consumption of Shanghai's urban rail transit was about 2.05 billion kWh. It can be seen that the annual electricity consumption of urban rail transit is huge and the growth rate is alarming, so energy conservation work is imminent.
[0004] The main energy-saving and consumption-reducing measures currently adopted by rail transit are: 1. Energy-saving of vehicle attributes: including the selection of vehicle models with large transport capacity, light weight, and low streamline resistance; 2. Energy-saving of line selection: including optimizing curve radius, optimizing line energy-saving slope, optimizing longitudinal slope design, etc.; 3. Energy-saving of power supply system: including reasonable design of substation location to reduce line loss, etc.; 4. Energy-saving of energy storage device: such as supercapacitor, flywheel energy storage, etc.; 5. Energy-saving of vehicle operation: including optimizing train ATO speed curve, adjusting operation diagram, etc. The above measures 1, 2, and 3 have been considered in the subway design stage. The cost of measure 4 is expensive and the application technology is immature. Measure 5 only needs to make minor adjustments to the operation of the rail transit system to reduce energy consumption, which is low-cost and simple. However, in the existing research, because the ATO speed curve and the operation diagram belong to two different disciplines, they are often studied separately as two independent fields. This has led to the failure to analyze the models of the two fields clearly at the same time in the initial exploration of integrated energy-saving that combines the two, and there is an inaccurate integrated energy-saving model, which cannot truly achieve the effect of integrated energy-saving. Summary of the invention
[0005] The purpose of the present invention is to provide a train operation energy-saving method based on dynamic programming and time allocation, which can not only ensure the punctual operation of trains, but also be easy to promote and implement, so as to achieve the goals of comprehensive energy saving and consumption reduction and intelligent optimization management of operations.
[0006] The technical solution to achieve the purpose of the present invention is: a train operation energy-saving method based on dynamic programming and time allocation, comprising the following steps:
[0007] Step S1: Based on the force analysis of the train running process and combined with the principle of dynamic programming, a multi-stage calculation model of train traction energy consumption is established;
[0008] Step S2: taking the weight of stage energy consumption and time as the optimization objective function, using the binary search method to search for the interval operation curve of the fixed time;
[0009] Step S3: inputting the fixed time interval operation curve obtained in step S2 into the multi-stage calculation model of traction energy consumption established in step S1 to obtain the corresponding traction energy consumption at different operation times in each interval;
[0010] Step S4: fitting the time energy consumption inverse proportional characteristic curves of different intervals based on the least squares method;
[0011] Step S5: Calculate the shortest running time of each interval, compare it with the planned running time to obtain the redundant time T, discretize the redundant time, and divide the redundant time into n segments with equal intervals, each segment is a time unit ΔT;
[0012] Step S6: Establishing a multi-interval operation curve collaborative optimization model based on time allocation;
[0013] Step S7: take a time unit and pre-allocate it to all intervals;
[0014] Step S8: Compare the energy consumption of the new running time and the original running time in each interval, and finally allocate the time unit to the interval with the largest energy consumption change;
[0015] Step S9: Update the interval running time and redundant time. If the redundant time T has been fully allocated, the allocation is completed and the process goes to step S10. Otherwise, the process jumps to step S7.
[0016] Step S10: Input the redundant time allocation result in step S9 into the ATS system and compile a new energy-saving operation diagram.
[0017] Compared with the prior art, the present invention has the following significant advantages: (1) it can optimize and adjust the train operation curve and operation diagram, which can not only meet the requirements of passenger comfort and ensure the punctual operation of the train, but also reduce the traction energy consumption of the train; (2) it is simple and easy to implement, and is conducive to achieving the goals of comprehensive energy saving and consumption reduction and intelligent operation optimization management. The experimental simulation conducted on the software simulation platform achieved an energy saving effect of 10.15%. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 It is a flow chart of the train operation energy-saving method based on dynamic programming and time allocation of the present invention.
[0019] Figure 2 It is a schematic diagram of a multi-stage train operation model constructed based on dynamic programming in the present invention.
[0020] Figure 3 It is a schematic diagram of the process of solving the operation curve by dynamic programming in the present invention.
[0021] Figure 4 It is a schematic diagram of the inverse proportional characteristic curve of time energy consumption of four intervals randomly selected in the present invention.
[0022] Figure 5 It is a schematic diagram of the multi-interval running time allocation process in the present invention.
[0023] Figure 6 It is a schematic diagram of the optimized operation curve of a certain section in the present invention and the data before and after the line up optimization. DETAILED DESCRIPTION
[0024] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments to enable a better understanding of the functions and features of the present invention. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.
[0025] Combination Figure 1 , a train operation energy-saving method based on dynamic programming and time allocation according to an embodiment of the present invention comprises the following steps:
[0026] Step S1: Based on the force analysis of the train running process and combined with the principle of dynamic programming, a multi-stage calculation model of train traction energy consumption is established;
[0027] Combination Figure 2 The multi-stage train operation model shown in the figure is constructed based on the dynamic programming principle, and the multi-stage calculation model of train traction energy consumption specifically includes:
[0028] Step 101: Using Δs as the segment spacing, divide the entire interval into many molecular intervals at equal intervals, sk is one of the subintervals, assuming that in the subinterval s k Within the interval s, the train acceleration is constant, and the slope and curve values of the line remain unchanged. k Inside
[0029]
[0030]
[0031] In the formula, v k is the train in subinterval s k The starting speed within; v k+1 is the train in subinterval s k Terminal velocity within a k,k+1 is the train in subinterval s k The acceleration of the internal operation; t k,k+1 is the train in subinterval s k The time taken to run within
[0032] Step 102: If the train is in sub-interval s k If the internal coasting is for:
[0033]
[0034] Where ρ is the rotation factor of the train mass; g is the gravitational acceleration of the train, in m / s 2 ;ω 0 is the unit basic resistance of the train, in kN / N; ω i is the unit slope resistance of the train, which is positive for uphill and negative for downhill, and the unit is kN / N; ω r is the unit curve resistance of the train, in kN / N;
[0035] Step 103: If Then in the interval s k The traction force of the train plays a role, and the traction energy consumption of the train for:
[0036]
[0037] In the formula, is the train in subinterval s k The traction energy consumed during operation is expressed in joules, and M is the mass of the train in tons;
[0038] Step 104: If Then in the interval s k The internal train braking force comes into play. Assuming that the energy consumption of the braking resistor is zero, the braking energy feedback of the train is:
[0039]
[0040] In the formula, is the train in subinterval s k The braking energy feedback during internal operation is expressed in joules;
[0041] Step 105: If Then in the interval s k The train is coasting, and the energy consumption of the train is:
[0042]
[0043] In the formula, is the train in subinterval s k Energy consumption of idling during internal operation;
[0044] Step 106: T is the time taken by the train to run in the entire section, in seconds; E is the traction energy consumption of the train running in the entire section, in joules; the calculation formulas of T and E are as follows:
[0045]
[0046]
[0047] Step S2: taking the weight of stage energy consumption and time as the optimization objective function, using the binary search method to search for the interval operation curve of the fixed time;
[0048] In order to achieve a balance between stage time and energy consumption, the energy consumption factor μ is taken, and the objective function f k,k+1 The weighting of operating energy consumption and time is adopted, namely:
[0049]
[0050] In the above formula, μ is the running time t of the stage in the objective function k,k+1 Relative to stage traction energy consumption The weight of μ is higher, the stage running time t k,k+1 The larger the weight is, the longer the train interval running time is and the lower the interval energy consumption is. Conversely, the smaller the weight is, the shorter the train interval running time is and the higher the interval energy consumption is.
[0051] Combination Figure 3 The flowchart of solving the operation curve based on dynamic programming is shown in the figure. The binary search of the fixed time interval operation curve specifically includes:
[0052] Step 201: Set the objective function f k,k+1 Rewrite it as follows:
[0053]
[0054] In the formula, α is a parameter in the interval (0, 1), e k,k+1 is the subinterval s k Traction energy consumption of the train, t k,k+1 is the subinterval s k The running time of the trains;
[0055] Step 202: Set and input the interval running time T i ;
[0056] Step 203: Divide the train running section into n-1 parts at equal intervals of Δs, each part is a sub-section, and there are n endpoints in total. The speed of each endpoint is recorded as a speed point. When dividing, the curve and slope between any two speed points are constant. If the change point of the slope and the curve is just between the two speed points, it is cost-effective according to the principle of proximity.
[0057] Step 204: according to the shortest running time strategy (maximum traction-cruise-maximum braking) when the train runs in the section, a speed point set under the strategy is obtained;
[0058] Step 205: Using Δv as the speed segment spacing, discretize the speed points of each stage under the shortest running time strategy, record the set of speed points that can be selected in each stage as the speed state space, and record the points that can be selected in the speed state space as the speed state points. The minimum value of the speed state points of each stage is 0, and the maximum value is the speed point of the stage under the shortest running time strategy;
[0059] Step 206: Starting from the n-1th subinterval, v n = 0, there is only one speed state point, let the objective function f of the n-1th stage n-1,n =0;
[0060] Step 207: Stage n-1, based on Bellman's principle
[0061]
[0062] In the formula, is the optimal index at the n-1th speed state point; is the optimal index at the nth speed state point. Since it is the end point, this index is marked as 0; one step forward, we get And so on;
[0063] Step 208: In the n-2th stage, let the speed state point number of this stage be j, j = 1, 2, 3, ... k, that is, there are k speed points that can be selected in this stage, then
[0064]
[0065] Calculate forward iteratively until n=1;
[0066] Step 209: After the iterative calculation is completed, the speed points, total energy consumption e_sum, and total running time time_sum of each stage in the optimization process are counted;
[0067] Step 210: Compare the total running time time_sum with the input interval running time T i If time_sum is not equal to T i , then the binary method is used to adjust the size of α, α∈(0,1), and steps 206 to 209 are repeated until |T i -time_sum|<δ, δ is the set threshold, the algorithm ends here, and the interval operation curve under fixed time is obtained.
[0068] Step S3: inputting the fixed time interval operation curve obtained in step S2 into the multi-stage calculation model of traction energy consumption established in step S1 to obtain the corresponding traction energy consumption at different operation times in each interval;
[0069] Step S4: Based on the least squares method, the time energy consumption inverse proportional characteristic curves of different intervals are fitted; Figure 4 As shown, the time-energy consumption inverse proportional characteristic curve means that the train has different running times in a certain section, and each running time has a corresponding minimum traction energy consumption. As the running time increases, the corresponding minimum traction energy consumption gradually decreases, and the relationship between the running time and the minimum traction energy consumption is inversely proportional, that is, each section has a unique time-energy consumption inverse proportional curve.
[0070] Step S5: Calculate the shortest running time of each interval, compare it with the planned running time to obtain the redundant time T, discretize the redundant time, and divide the redundant time into n segments with equal intervals, each segment is a time unit ΔT;
[0071] The shortest running time of each section refers to the running time of each section when the train ATO runs in the maximum traction-cruise-maximum braking mode;
[0072] The redundant time T refers to the difference between the sum of the running time of each section of the entire line specified in the planned timetable and the sum of the shortest running time of each section.
[0073] Step S6: Establishing a multi-interval operation curve collaborative optimization model based on time allocation;
[0074] Step S7: take a time unit and pre-allocate it to all intervals;
[0075] Step S8: Compare the energy consumption of the new running time and the original running time in each interval, and finally allocate the time unit to the interval with the largest energy consumption change;
[0076] Step S9: Update the interval running time and redundant time. If the redundant time T has been fully allocated, the allocation is completed and the process goes to step S10. Otherwise, the process jumps to step S7.
[0077] Combination Figure 5 Steps S5 to S9 specifically include:
[0078] Step 601: Taking the total traction energy consumption of the train going up or down as the optimization target, the objective function is shown in the following formula:
[0079]
[0080] In the above formula, E(T) represents the total traction power consumption of the train going up or down, in kw·h; N represents the number of line operation sections, which is the total number of stations minus 1; E j (T j ) indicates that the train runs in the jth section and the running time is T j The traction power consumption.
[0081] Step 602: The total running time of the train must meet the constraint of the total running time of the line plan to ensure that the train departure interval is not affected by the redundant time allocation timeout. The total running time constraint is shown in the following formula.
[0082]
[0083] T j represents the running time of the train in the jth section, x j represents the train’s stop time at the jth station, T total It represents the total running time of the train after time allocation, T plan It indicates the total planned operating time of the line. Considering that the passenger flow at different stations varies greatly, the stop time at each station is also different, usually between 25 seconds and 45 seconds.
[0084] Step 603: Calculate the redundant time T for the uplink, and calculate the sum of the shortest running time T for each uplink section of the train. min , the sum of the planned running time of each uplink section is T plan , T plan Subtract T min The result is the uplink redundant time T;
[0085] Step 604: divide the redundant time into n equal parts, each part is a time unit ΔT, ΔT = T / n;
[0086] Step 605: pre-allocate ΔT to all uplink intervals, and combine the time energy consumption inverse proportional characteristic curve of each interval to obtain the energy consumption change ΔE of each interval i ;
[0087] Step 606: Compare ΔE i Finally, the time unit ΔT is allocated to the interval with the largest energy consumption change, and the running time T of this interval is updated. i =T i +ΔT, update redundancy time T = T-ΔT;
[0088] Step 607: If the running time of the i-th interval reaches the maximum running time of the interval, no redundant time is allocated to the interval;
[0089] Step 608: If the redundant time T is not zero, repeat steps 605 to 607, otherwise, the algorithm ends;
[0090] Step S10: Input the redundant time allocation result in step S9 into the ATS (Automatic Train Supervision) system to compile a new energy-saving operation diagram, such as Figure 6 shown.
[0091] The method of the present invention is used to simulate actual line data of a domestic subway line. The simulation results show that the total traction energy consumption of each section of the train going up is reduced by 10.15%, achieving a good energy-saving effect.
[0092] In summary, the method of the present invention uses software simulation to explore a new way for energy-saving optimization at the overall level of the line. It can not only meet the requirements of passenger comfort and ensure the punctual operation of the train, but also reduce the traction energy consumption of the train. The method is simple and easy to implement and easy to promote and implement. At present, the test simulation using actual line data on the software simulation platform has achieved an energy-saving effect of 10.15%. In the later stage, it will be tested on the actual line. If a good optimization effect is achieved, it can be promoted to the national subway to achieve the goal of comprehensive energy saving and consumption reduction and intelligent optimization management of operations.
Claims
1. A train operation energy-saving method based on dynamic programming and time allocation, characterized in that: The following steps are involved: Step S1: Based on the force analysis of the train running process and combined with the principle of dynamic programming, a multi-stage calculation model of train traction energy consumption is established; Step S2: taking the weight of stage energy consumption and time as the optimization objective function, using the binary search method to search for the interval operation curve of the fixed time; Step S3: inputting the fixed time interval operation curve obtained in step S2 into the multi-stage calculation model of traction energy consumption established in step S1 to obtain the corresponding traction energy consumption at different operation times in each interval; Step S4: fitting the time energy consumption inverse proportional characteristic curves of different intervals based on the least squares method; Step S5: Calculate the shortest running time of each interval, compare it with the planned running time to obtain the redundant time T, discretize the redundant time, and divide the redundant time into n segments with equal intervals, each segment is a time unit ΔT; Step S6: Establishing a multi-interval operation curve collaborative optimization model based on time allocation; Step S7: take a time unit and pre-allocate it to all intervals; Step S8: Compare the energy consumption of the new running time and the original running time in each interval, and finally allocate the time unit to the interval with the largest energy consumption change; Step S9: Update the interval running time and redundant time. If the redundant time T has been fully allocated, the allocation is completed and the process goes to step S10. Otherwise, the process jumps to step S7. Step S10: input the redundant time allocation result in step S9 into the ATS system to compile a new energy-saving operation diagram; The forces applied during the train operation process in step S1 include train traction, braking force, slope resistance, curve resistance and basic resistance; The establishment of the multi-stage calculation model for train traction energy consumption in step S1 is a single-point model based on position, which is as follows: Step 101: Using Δs as the segment spacing, divide the entire interval into many molecular intervals at equal intervals, s k is one of the subintervals, assuming that in the subinterval s k Within the interval s, the train acceleration is constant, and the slope and curve values of the line remain unchanged. k Inside In the formula, v k is the train in subinterval s k The starting speed within; v k+1 is the train in subinterval s k Terminal velocity within a k,k+1 is the train in subinterval s k The acceleration of the internal operation; t k,k+1 is the train in subinterval s k The time taken to run within Step 102: If the train is in sub-interval s k If the internal coasting is for: Where ρ is the rotation factor of the train mass; g is the gravitational acceleration of the train, in m / s 2 ; ω0 is the unit basic resistance of the train, in kN / N; ω i is the unit slope resistance of the train, which is positive for uphill and negative for downhill, and the unit is kN / N; ω r is the unit curve resistance of the train, in kN / N; Step 103: If Then in the interval s k The traction force of the train plays a role, and the traction energy consumption of the train for: In the formula, is the train in subinterval s k The traction energy consumed during operation is expressed in joules; M is the mass of the train in tons; Step 104: If Then in the interval s k The internal train braking force comes into play. Assuming that the energy consumption of the braking resistor is zero, the braking energy feedback of the train is: In the formula, is the train in subinterval s k The braking energy feedback during internal operation is expressed in joules; Step 105: If Then in the interval s k The train is coasting, and the energy consumption of the train is: In the formula, is the train in subinterval s k Energy consumption of idling during internal operation; Step 106: T is the time taken by the train to run in the entire section, in seconds; E is the traction energy consumption of the train running in the entire section, in joules; the calculation formulas of T and E are as follows: The optimization objective function described in step S2 using the weighted energy consumption and time of each stage is as follows: In order to achieve a balance between stage time and energy consumption, the energy consumption factor μ is taken, and the objective function f k,k+1 The weighting of phase energy consumption and time is adopted, namely: In the above formula, μ is the running time t of the stage in the objective function k,k+1 Relative to stage traction energy consumption The weight of μ is higher, the stage running time t k,k+1 The larger the weight is, the longer the train interval running time is and the lower the interval energy consumption is. On the contrary, the smaller the weight is, the shorter the train interval running time is and the higher the interval energy consumption is. Step S2 uses the binary search method to search for a fixed time interval operation curve, as follows: Step 201: Set the objective function f k,k+1 Rewrite it as follows: In the formula, α is a parameter in the interval (0, 1), e k,k+1 is the subinterval s k Traction energy consumption of the train, t k,k+1 is the subinterval s k The running time of the trains; Step 202: Set and input the interval running time T i ; Step 203: Divide the train running section into n-1 parts at equal intervals of Δs, each part is a sub-section, and there are n endpoints in total. The speed of each endpoint is recorded as a speed point. When dividing, the curve and slope between any two speed points are constant. If the change point of the slope and the curve is just between the two speed points, it is cost-effective according to the principle of proximity. Step 204: according to the shortest running time strategy when the train runs in the section, a set of speed points under the strategy is obtained; Step 205: Using Δv as the speed segment spacing, discretize the speed points of each stage under the shortest running time strategy, record the set of speed points that can be selected in each stage as the speed state space, and record the points that can be selected in the speed state space as the speed state points. The minimum value of the speed state points of each stage is 0, and the maximum value is the speed point of the stage under the shortest running time strategy; Step 206: Starting from the n-1th subinterval, v n = 0, there is only one speed state point, let the objective function f of the n-1th stage n-1,n =0; Step 207: Stage n-1, based on Bellman's principle In the formula, is the optimal index at the n-1th speed state point; is the optimal index at the nth speed state point. Since it is the end point, this index is marked as 0; one step forward, we get And so on; Step 208: In the n-2th stage, let the speed state point number of this stage be j, j = 1, 2, 3, ... k, that is, there are k speed points that can be selected in this stage, then Iterate forward until n=1; Step 209: After the iterative calculation is completed, the speed points, total energy consumption e_sum, and total running time time_sum of each stage in the optimization process are counted; Step 210: Compare the total running time time_sum with the input interval running time T i If time_sum is not equal to T i , then the binary method is used to adjust the size of α, α∈(0,1), and steps 206 to 209 are repeated until |T i -time_sum|<δ, δ is the set threshold, the algorithm ends here, and the interval operation curve under fixed time is obtained.
2. The train operation energy-saving method based on dynamic programming and time allocation according to claim 1 is characterized in that: The time-energy consumption inverse proportional characteristic curve in step S4 means that the train has different running times in a certain section, and each running time has a corresponding minimum traction energy consumption. As the running time increases, the corresponding minimum traction energy consumption gradually decreases. The relationship between the running time and the minimum traction energy consumption is inversely proportional, that is, each section has a unique time-energy consumption inverse proportional curve.
3. The train operation energy-saving method based on dynamic programming and time allocation according to claim 1 is characterized in that: The shortest running time of each section in step S5 refers to the running time of each section when the train ATO runs in the maximum traction-cruise-maximum braking mode; The redundant time T refers to the difference between the sum of the running time of each section of the entire line specified in the planned timetable and the sum of the shortest running time of each section.
4. The train operation energy-saving method based on dynamic programming and time allocation according to claim 1 is characterized in that: Steps S6 to S9 specifically include: Step 601: Taking the total traction energy consumption of the train going up or down as the optimization target, the objective function is shown in the following formula: In the above formula, E(T) represents the total traction power consumption of the train going up or down, in kw·h; N represents the number of line operation sections, which is the total number of stations minus 1; E j (T j ) indicates that the train runs in the jth section and the running time is T j Traction power consumption at Step 602: The total running time of the train must meet the constraint of the total running time of the line plan to ensure that the train departure interval is not affected by the redundant time allocation timeout. The total running time constraint is as follows: In the above formula, T j represents the running time of the train in the jth section, x j represents the train’s stop time at the jth station, T total It represents the total running time of the train after time allocation, T plan Indicates the total planned operating time of the line; Step 603: Calculate the redundant time T for the uplink, and calculate the sum of the shortest running time T for each uplink section of the train. min , the sum of the planned running time of each uplink section is T plan , T plan Subtract T min The result is the uplink redundant time T; Step 604: divide the redundant time into n equal parts, each part is a time unit ΔT, ΔT=T / n; Step 605: pre-allocate ΔT to all uplink intervals, and combine the time energy consumption inverse proportional characteristic curve of each interval to obtain the energy consumption change ΔE of each interval i ; Step 606: Compare ΔE i Finally, the time unit ΔT is allocated to the interval with the largest energy consumption change, and the running time T of this interval is updated. i =T i +ΔT, update redundancy time T = T-ΔT; Step 607: If the running time of the i-th interval reaches the maximum running time of the interval, no redundant time is allocated to the interval; Step 608: If the redundant time T is not zero, repeat steps 605 to 607; otherwise, end.
