Intelligent control method for rail transit traction power supply system

The intelligent control method addresses inefficiencies in rail transit traction power supply by optimizing power distribution and regenerative energy utilization, enhancing system reliability and stability through real-time data processing and closed-loop adjustments.

CN119093309APending Publication Date: 2024-12-06CRRC QINGDAO SIFANG ROLLING STOCK RESEARCH INSTITUTE CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411205464.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing rail transit traction power supply systems face issues with regenerative braking energy recovery, uncontrollable DC grid voltage, inefficiencies, and power supply imbalances, leading to energy waste, safety risks, and suboptimal train operation efficiency.

Method used

An intelligent control method that includes real-time data collection, preprocessing, optimization control using nonlinear dynamic programming, and closed-loop system adjustment to maximize regenerative energy utilization, minimize transmission losses, and stabilize voltage.

Benefits of technology

Improves energy utilization efficiency, reduces operating costs, enhances system reliability and stability by optimizing power distribution and suppressing circulating currents, ensuring stable and efficient power supply.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093309A_ABST
    Figure CN119093309A_ABST
Patent Text Reader

Abstract

The invention relates to an intelligent control method for a rail transit traction power supply system. The intelligent control method comprises the following steps of data collection, wherein power monitoring system data, signal system data and train operation data are collected in real time; data processing: preprocessing the collected real-time data, including delay compensation and abnormal data correction; optimization control: based on the compensated and corrected data, performing operation by using an intelligent algorithm, and determining a power distribution scheme of the direct current network; instruction issuing: converting the optimized power distribution scheme into a control instruction, and sending the control instruction to each converter station control center through the Ethernet; and system adjustment: each converter station control center adjusts the terminal voltage of the station according to the received control instruction to form system closed-loop operation. The technical problems that in the prior art, regenerative braking energy of a rail transit traction power supply system cannot be efficiently recycled, the direct current network voltage is uncontrollable, and the efficiency of the power supply system is low are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of rail transit, and in particular relates to an intelligent control method for a rail transit traction power supply system. Background Art

[0002] Rail trains require DC power supply. Currently, urban rail transit power grids usually use multi-pulse rectifiers to achieve the conversion of 35kV or 10kV medium voltage grids to 750V or 1500V DC grids to power trains. This rectifier power supply method is widely used, mature, and low-cost. It also has the advantages of high reliability, but it also has the following problems:

[0003] 1. Due to the unidirectional working characteristics of diode rectification, the regenerative braking energy of line trains cannot be recovered, resulting in an increase in DC voltage, which endangers the safety of driving and power supply equipment. To solve related problems, there are two commonly used solutions in the industry. One is to equip a braking resistor to consume regenerative energy, but this solution will cause energy waste and increase the burden on tunnel ventilation; the second solution is to use an energy feedback device to absorb regenerative energy, but the energy feedback device can only work in the inverter state, and the equipment utilization rate is low;

[0004] 2. The DC grid voltage is uncontrollable and changes with the power consumption of the train. The greater the power of the train during traction, the lower the DC grid voltage, which can be as low as about 1300V;

[0005] 3. The rectifier unit is a passive power supply. The train distribution on the line changes in real time, the power changes in real time, the power load of the entire line is unbalanced, the power supply is also unbalanced, and the system efficiency cannot reach the optimal level;

[0006] 4. The DC voltage fluctuates greatly, with the maximum difference of about 600V, which brings many restrictions to train design and train control, and the train operation efficiency cannot reach the optimal level.

[0007] With the widespread application of power electronics and the improvement of reliability, the above problems can be optimized by replacing some rectifier units with bidirectional power supply devices. Bidirectional power supply devices are four-quadrant converters that can control the power supply or feedback mode and control the DC voltage and power. Placing bidirectional power supply devices in each substation can reduce line voltage fluctuations. The bidirectional power supply control targets of each substation can only be set uniformly, and the above problems cannot be completely solved.

[0008] With the development of information technology, the operating data of various systems have been mastered more and more comprehensively on the ground. In order to manage and control the power supply at the system level, Sifang Institute proposed the concept of energy management and control system. By obtaining power supply and power consumption information, the input and output power of the power supply and the network voltage are controlled to realize active power supply, achieve the lowest system energy consumption and the highest efficiency.

[0009] At present, relevant technical research and application of bidirectional converters have been carried out both at home and abroad. The HESOP reversible substation developed by Alstom completely replaces the traditional rectifier unit power supply system with a bidirectional converter. It has been installed and put into use in many countries around the world. Japan's Tsukuba Express has also applied bidirectional converters on the entire line. However, there are few cases of achieving system-level energy control. Among them, Siemens has achieved station energy control abroad, and Toshiba and Hitachi have also achieved coordinated control of vehicle traction power supply. At present, only Zhuzhou Institute and Sifang Institute have proposed the concept of coordinated energy control of the entire line based on power consumption information. Zhuzhou Institute conducted an energy management system verification test "Urban Rail Transit Intelligent Traction Power Supply System" on Ningbo Metro Line 4. The system achieved a total power saving of 0.767%, which is about 5% energy saving compared to the traditional rectifier unit plus energy feed system power supply method, and the network voltage fluctuation of the entire line is reduced by about 40% compared with the traditional power supply system, but it has not yet been implemented in engineering and put into operation. Summary of the invention

[0010] Aiming at the deficiencies existing in the related technologies, the present invention solves the technical problems in the prior art that the regenerative braking energy of the rail transit traction power supply system cannot be recovered, the DC grid voltage is uncontrollable, and the power supply system is inefficient. The present invention provides an intelligent control method for the rail transit traction power supply system.

[0011] In a possible implementation, a rail transit traction power supply system intelligent control method includes the following steps:

[0012] S10, data collection: real-time collection of power monitoring system data, signal system data and train operation data;

[0013] S20, data processing: pre-processing the collected real-time data, including delay compensation and abnormal data correction;

[0014] S30, optimization control: based on the compensated and corrected data, use intelligent algorithms to calculate and determine the power distribution plan of the DC network;

[0015] S40, issuing instructions: converting the optimized power allocation scheme into control instructions, and sending them to the control center of each converter station via Ethernet;

[0016] S50, system adjustment: each converter station control center adjusts the station terminal voltage according to the received control instructions to form a closed-loop operation of the system.

[0017] In a possible implementation, in step S20, delay compensation includes: by comparing timestamps of different data sources, synchronizing the data so that the data are aligned under the same time reference.

[0018] In a possible implementation, abnormal data correction includes: automatically filtering and correcting abnormal data points that exceed a normal range by setting a threshold range.

[0019] In a possible implementation, the optimization control step includes:

[0020] Using nonlinear dynamic programming algorithm, combined with the physical constraints and objective function of the system, the power distribution of the DC grid is optimized;

[0021] Among them, the nonlinear dynamic programming algorithm for optimal control is described by the following formula:

[0022] DC side power flow calculation: YU=I(1); Y is the node admittance matrix, U is the node voltage vector, and I is the injected current vector;

[0023] The AC side power flow calculation is:

[0024]

[0025] Among them, P s is the power generation, V s is the voltage, G ij and B ij are the real and imaginary parts of the admittance matrix, θ ij is the voltage phase angle difference;

[0026] The mathematical model of the interface between the AC side and the DC side is:

[0027]

[0028] φ is the power factor angle, T i is the traction transformer ratio, I dc The DC port of the traction substation supplies current.

[0029] In a possible implementation, in step S30, the control algorithm is optimized to achieve the following objectives:

[0030] Maximize the utilization rate of regenerative energy, including: adjusting the voltage of the converter station and reducing the energy lost in the braking resistor during the vehicle braking phase;

[0031] The calculation formula for brake resistance loss is:

[0032]

[0033] Assume that the vehicle power during deceleration is P i >0, the voltage threshold for closing the brake resistor is V br , considering r f and L f The impact of the vehicle on the traction grid side voltage threshold can be set to Vbr -δ br ; Among them, δ br Represents the voltage drop on the equivalent resistance of the vehicle network; P br (u i ) is the braking resistor loss, u i is the voltage, V ub is the voltage threshold, δ br is the pressure drop, P i is power, r br is the braking resistor;

[0034] Minimize the power line transmission loss, including: controlling the voltage of the converter station to reduce the loss of power during transmission; line loss is the difference between power generation and power consumption, calculated by the following formula:

[0035]

[0036] Among them, P loss (u i ) is the line loss, is the total AC and DC power generation, is the total AC and DC loads;

[0037] Suppressing the circulating current in the station, including: when the bidirectional energy feeding device is in the state of feeding power to the grid and the voltage is lower than the open circuit voltage of the rectifier, the energy will be dissipated by the circulating current between the bidirectional energy feeding device and the rectifier. When the open circuit voltage of the rectifier is U oc When , the calculated power of the circulating current in the station is:

[0038]

[0039] And, large bilateral voltage stabilization control, including: when a rectifier is detected to be down, the adjacent node converter station will provide voltage support and ignore the optimization operation mode; assuming that the rated voltage of site i is At this moment, the measured voltage is u i , then the voltage setting value of the adjacent node j, h converter station is adjusted to:

[0040]

[0041] Among them, K j and K h is the droop ratio, which is determined by the capacity of stations j and h and the distance to station i.

[0042] In a possible implementation, the nonlinear dynamic programming overall modeling of the rail transit power supply system is expressed as:

[0043]

[0044] Among them, the sequential quadratic programming (SQP) method is used to solve the above equation according to the system characteristics. When solving, the tram power is kept constant, and the optimal voltage setting value of each substation can be obtained.

[0045] In one possible implementation, determining Whether it is necessary to make adjustments according to formula (7); if no adjustment is required, then fine-tune according to formula (9) based on the rectification droop characteristics and set it as the target voltage value;

[0046]

[0047] In the formula, K d is the droop slope, is the rated current of device i, I i It is the actual measured current.

[0048] In a possible implementation, the instruction issuing step includes:

[0049] The optimized power allocation scheme is converted into specific control instructions and sent to the control center of each converter station via Ethernet for execution.

[0050] In a possible implementation, the system adjustment step includes:

[0051] The control center of each converter station receives and executes control instructions to adjust the voltage at the station end.

[0052] Based on the above technical scheme, the intelligent control method of the rail transit traction power supply system of the present invention collects and pre-processes power monitoring, signal system and train operation data in real time through the system; uses intelligent algorithms to perform optimization control, determines the power distribution plan of the DC network, maximizes the utilization rate of regenerated energy, minimizes transmission losses, and suppresses circulating currents within the station; the optimized plan is converted into control instructions, which are sent to the control center of each converter station via Ethernet, adjusts the station end voltage, forms a closed-loop operation, improves energy utilization efficiency, reduces operating costs, and enhances the reliability and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0054] Figure 1 This is an overall structural diagram of an intelligent energy management system applied to an intelligent control method for a rail transit traction power supply system according to an embodiment of the present invention;

[0055] Figure 2 It is the core functional module of the intelligent energy management system;

[0056] Figure 3 An illustrative example of data fitting;

[0057] Figure 4 It is the equivalent circuit of the substation and the vehicle;

[0058] Figure 5 It is the equivalent circuit of the rail voltage algorithm of Jinan Metro Line 3 Phase II network;

[0059] Figure 6 This is the flow chart of static DC side power flow calculation;

[0060] Figure 7 is the equivalent circuit diagram of the AC side;

[0061] Figure 8 Detailed equivalent circuit for the vehicle;

[0062] Fig. 9 This is a schematic diagram of the circulation within the station;

[0063] Fig.10 The present invention is a flow chart of the intelligent control method for the rail transit traction power supply system. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0065] In the description of the present invention, it should be understood that the terms "center", "lateral", "longitudinal", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0066] The terms "first", "second", and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Thus, a feature defined as "first", "second", and "third" may explicitly or implicitly include one or more of the features.

[0067] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0068] In order to solve the technical problems in the prior art that the regenerative braking energy of the rail transit traction power supply system cannot be recovered, the DC grid voltage is uncontrollable, and the power supply system is inefficient, the present application provides an intelligent control method for the rail transit traction power supply system.

[0069] See also Fig.10 In a possible implementation manner, the intelligent control method of the rail transit traction power supply system includes the following steps:

[0070] S10, data collection: real-time collection of power monitoring system data, signal system data and train operation data;

[0071] S20, data processing: pre-processing the collected real-time data, including delay compensation and abnormal data correction;

[0072] S30, optimization control: based on the compensated and corrected data, use intelligent algorithms to calculate and determine the power distribution plan of the DC network;

[0073] S40, issuing instructions: converting the optimized power allocation scheme into control instructions, and sending them to the control center of each converter station via Ethernet;

[0074] S50, system adjustment: each converter station control center adjusts the station terminal voltage according to the received control instructions to form a closed-loop operation of the system.

[0075] In this implementation, first, the relevant data of the power monitoring system, signal system and train operation are acquired in real time through the data acquisition step S10; these data are subjected to delay compensation and abnormal data correction through the data processing step S20 to ensure the accuracy and real-time performance of the data; the processed data are subjected to optimization control step S30, and the intelligent algorithm is used for analysis and calculation to generate a power distribution plan for the DC network; next, the optimized plan is converted into a specific control instruction through the instruction issuance step S40, and sent to each converter station control center through Ethernet; finally, in step S50, the converter station adjusts the voltage at each station end according to the received control instruction to complete the closed-loop control of the system.

[0076] The intelligent control method of the above scheme can collect and process a variety of system data in real time, improving the accuracy and real-time performance of the data; through optimized control and intelligent algorithms, the power of the DC network can be effectively distributed, energy waste can be reduced, and system efficiency can be improved; the instruction issuance and system adjustment links ensure the accurate execution of control instructions and maintain stable operation of the system; in addition, the method realizes automatic dynamic adjustment through closed-loop control, further improving the reliability and safety of the system.

[0077] In a possible implementation, in step S20, delay compensation includes: by comparing timestamps of different data sources, synchronizing the data so that the data are aligned under the same time reference.

[0078] In the above scheme, due to the time transmission deviation of different data sources, the system first aligns the timestamps of the data packets. The alignment criteria are as follows:

[0079] Vehicle data: Vehicle data with a time difference of less than 300 milliseconds is considered to be data of the same time section.

[0080] Vehicle data and bidirectional variable flow data: Data with a time difference of less than 30 milliseconds are regarded as data of the same time section.

[0081] Traction network current and vehicle power total current: By comparing the total current of the traction network power supply and the total current of the vehicle power supply, the data loss situation is judged. If the difference between the two is less than 500A, the data packet is complete, otherwise it will be discarded.

[0082] In this embodiment, delay compensation is achieved by comparing the timestamps of different data sources. First, the timestamps of each data source are obtained and compared. Then, the data is synchronized according to the comparison results so that all data are aligned under the same time reference, thereby eliminating the error caused by time delay.

[0083] This delay compensation method can effectively synchronize data from different data sources and ensure the accuracy of data processing and analysis. By synchronizing data, the error caused by time delay can be reduced and the overall performance and reliability of the system can be improved.

[0084] In a possible implementation, abnormal data correction includes: automatically filtering and correcting abnormal data points that exceed a normal range by setting a threshold range.

[0085] In this embodiment, abnormal data correction is achieved by setting a threshold range. First, a reasonable threshold range is set based on historical data and system characteristics. When real-time data exceeds the set threshold range, the system will automatically identify these abnormal data points and make corrections. The correction method can be to use the average of the normal data before and after or other data filling techniques to ensure the continuity and rationality of the data.

[0086] The abnormal data correction method can automatically filter and correct abnormal data points that exceed the normal range, ensuring the accuracy and reliability of the data. By setting the threshold range, abnormal situations can be discovered and handled in a timely manner, avoiding system misjudgments or failures caused by data anomalies, and improving the stability and security of the system.

[0087] In a possible implementation, abnormal data correction includes data verification:

[0088] Data stack storage: Data is stacked and stored in a cycle of 0.5 seconds, and individual data points may be lost during the process.

[0089] Smoothing Spline fitting: Smoothing Spline fitting is used to process the entire 10-second data. The missing data is compensated by the fitting function, and the influence of ripple and measurement error is filtered out.

[0090] In a possible implementation, the optimization control step includes:

[0091] Using nonlinear dynamic programming algorithm, combined with the physical constraints and objective function of the system, the power distribution of the DC grid is optimized;

[0092] Among them, the nonlinear dynamic programming algorithm for optimal control is described by the following formula:

[0093] DC side power flow calculation: YU=I(1); Y is the node admittance matrix, U is the node voltage vector, and I is the injected current vector;

[0094] The AC side power flow calculation is:

[0095]

[0096] Among them, P s is the power generation, V s is the voltage, G ij and B ij are the real and imaginary parts of the admittance matrix, θ ij is the voltage phase angle difference;

[0097] The mathematical model of the interface between the AC side and the DC side is:

[0098]

[0099] φ is the power factor angle, T i is the traction transformer ratio, I dc The DC port of the traction substation supplies current.

[0100] In this embodiment, the optimization control step uses a nonlinear dynamic programming algorithm to optimize the power distribution of the DC network. First, the relationship between the node voltage and current is determined by the DC side power flow calculation formula. Then, the power distribution scheme is optimized by the AC side power flow calculation formula, considering the physical constraints and objective function of the system. The interface part model of the AC side and the DC side further ensures the reasonable distribution of power factor and current.

[0101] In the above scheme, the AC / DC power flow model (modeling method) is used to analyze the voltage and power distribution of the system under different load conditions. According to the output of the power flow model, the voltage setting of the substation is adjusted to optimize the power transmission efficiency and system stability. The dynamic programming algorithm is used to find the best balance between multiple goals such as energy saving, voltage regulation, and fault response. Based on the results of dynamic programming, the system automatically adjusts the control parameters at different times to meet real-time operation requirements.

[0102] This optimization control method can make full use of the advantages of nonlinear dynamic programming algorithms, and achieve optimal distribution of DC network power on the basis of considering the system physical constraints and objective functions; improve the utilization efficiency of electric energy, reduce energy consumption, and ensure the stability and reliability of system operation.

[0103] In a possible implementation, in step S30, the control algorithm is optimized to achieve the following objectives:

[0104] Maximize the utilization rate of regenerative energy, including: adjusting the voltage of the converter station and reducing the energy lost in the braking resistor during the vehicle braking phase;

[0105] The calculation formula for brake resistance loss is:

[0106]

[0107] Assume that the vehicle power during deceleration is P i >0, the voltage threshold for closing the brake resistor is V br , considering r f and L f The impact of the vehicle on the traction grid side voltage threshold can be set to V br -δ br ; Among them, δ br Represents the voltage drop on the equivalent resistance of the vehicle network; P br (u i ) is the braking resistor loss, ui is the voltage, V ub is the voltage threshold, δ br is the pressure drop, P i is power, r br is the braking resistor;

[0108] Minimize the power line transmission loss, including: controlling the voltage of the converter station to reduce the loss of power during transmission; line loss is the difference between power generation and power consumption, calculated by the following formula:

[0109]

[0110] Among them, P loss (u i ) is the line loss, is the total AC and DC power generation, is the total AC and DC loads;

[0111] Suppressing the circulating current in the station, including: when the bidirectional energy feeding device is in the state of feeding power to the grid and the voltage is lower than the open circuit voltage of the rectifier, the energy will be dissipated by the circulating current between the bidirectional energy feeding device and the rectifier. When the open circuit voltage of the rectifier is U oc When , the calculated power of the circulating current in the station is:

[0112]

[0113] And, large bilateral voltage stabilization control, including: when a rectifier is detected to be down, the adjacent node converter station will provide voltage support and ignore the optimization operation mode; assuming that the rated voltage of site i is At this moment, the measured voltage is u i , then the voltage setting value of the adjacent node j, h converter station is adjusted to:

[0114]

[0115] Among them, K j and K h is the droop ratio, which is determined by the capacity of stations j and h and the distance to station i.

[0116] In this embodiment, the optimization control algorithm maximizes the utilization of regenerative energy by adjusting the voltage of the converter station, reduces the energy loss of the braking resistor, and optimizes the energy management of the vehicle braking phase by setting reasonable voltage thresholds and voltage drop parameters. In addition, by controlling the voltage of the converter station, the loss of electric energy during transmission is reduced, and by setting a reasonable line loss calculation formula, the loss of electric energy transmission is minimized. The method of suppressing the circulating current within the station dissipates excess energy and ensures system stability through the synergy of the bidirectional energy feedback device and the rectifier device. The large bilateral voltage stabilization control provides voltage support through adjacent nodes when the rectifier device is down to maintain the normal operation of the system.

[0117] This optimization control method can maximize the utilization of regenerative energy, reduce the loss of braking resistors, and achieve efficient energy management. At the same time, by controlling the voltage of the converter station and optimizing the line loss, the power transmission loss is reduced, and the energy efficiency and economy of the system are improved. The method of suppressing the circulating current in the station further improves the stability and reliability of the system. The large bilateral voltage stabilization control ensures that the system can still operate stably when the rectifier device is down, enhancing the robustness of the system.

[0118] In a possible implementation, the nonlinear dynamic programming overall modeling of the rail transit power supply system is expressed as:

[0119]

[0120] Among them, the sequential quadratic programming (SQP) method is used to solve the above equation according to the system characteristics. When solving, the tram power is kept constant, and the optimal voltage setting value of each substation can be obtained.

[0121] In this implementation, the overall nonlinear dynamic programming modeling introduces objective functions such as power loss, regenerative energy utilization, and circulating current suppression, and combines the physical constraints of the system to establish a complete optimization model. The optimization model is solved by the sequential quadratic programming method (SQP), and the optimal voltage setting value of each substation is obtained under the premise of ensuring the constant power of the tram, thereby achieving optimal control of the system.

[0122] This modeling method can comprehensively consider the various optimization objectives of the system and establish a complete nonlinear dynamic programming model. By solving the problem through the sequential quadratic programming method, the optimal control parameters of the system can be obtained while ensuring the constant power of the tram, so as to maximize the utilization of regenerative energy, minimize the power transmission loss, and optimize the circulation suppression effect. This method improves the operating efficiency and stability of the system.

[0123] In one possible implementation, determining Whether it is necessary to make adjustments according to formula (7); if no adjustment is required, then fine-tune according to formula (9) based on the rectification droop characteristics and set it as the target voltage value;

[0124]

[0125] Where K_d is the droop slope, I_i^* is the rated current of device i, and I_i is the actual measured current.

[0126] In this embodiment, the optimal voltage setting value calculated by formula (7) is It is necessary to determine whether to make voltage regulation control adjustments based on the power monitoring (PSCADA) instruction. If it is found that no further adjustment is required, the rectifier droop characteristic, that is, formula (9), is used to adjust the voltage regulation control. Fine-tune and finally set it as the target voltage value. This process ensures the rationality and effectiveness of the optimal voltage setting value.

[0127] By checking and adjusting the optimal voltage setting value, the accuracy and effectiveness of the target voltage value are ensured. The fine-tuning of the rectifier droop characteristics further optimizes the voltage setting and improves the stability and operating efficiency of the system.

[0128] In a possible implementation, the instruction issuing step includes:

[0129] The optimized power allocation scheme is converted into specific control instructions and sent to the control center of each converter station via Ethernet for execution.

[0130] In this implementation, the optimized power distribution scheme is processed and converted into specific control instructions. These control instructions are sent to each converter station control center via Ethernet. After receiving the instructions, each control center executes them and adjusts the corresponding voltage and power distribution to ensure the normal operation of the system.

[0131] In a possible implementation, the system adjustment step includes:

[0132] The control center of each converter station receives and executes control instructions to adjust the voltage at the station end.

[0133] In this implementation, after receiving the control instruction, each converter station control center adjusts the corresponding station terminal voltage according to the instruction content. By adjusting the station terminal voltage in real time, dynamic control and optimized operation of the system are achieved to ensure the reasonable distribution and stable transmission of voltage and power.

[0134] The intelligent control method of the rail transit traction power supply system according to one embodiment of the present invention is specifically described as follows:

[0135] This patent develops an intelligent control method for the energy management system. For the urban rail power supply system with a bidirectional energy feed device, it collects the operation data of the line vehicles and the power supply system, obtains the power consumption information, and determines the power distribution of the DC network according to the objective function and the system physical constraints, and converts it into the voltage command value of each substation and sends it down. As a top-level dispatching system, it optimizes the voltage and power output of the distributed traction power supply equipment to achieve the goals of active power supply, system energy saving, voltage stabilization and optimization of power supply quality. A system control method considering the influence of network transmission, data quality and physical system constraints is given.

[0136] Technical solutions for the software part:

[0137] 1. Overall process and core algorithm of urban rail intelligent energy control system

[0138] The intelligent energy management system is the top-level control system of the rail transit power supply system. Its overall process is as follows: Figure 1 This patent mainly focuses on the core algorithm of the energy management system, which includes two modules, namely the data processing module and the optimization control module. Figure 1 .

[0139] First, the rail transit measurement system collects power monitoring system data, signal system data, train operation data, etc. in real time through a fiber optic ring network.

[0140] The data is then sent via Ethernet to the computing server where the urban rail intelligent energy management and control system is deployed.

[0141] Then, the data processing module of the energy control system first pre-processes the real-time data collected by the rail transit measurement system, including delay compensation and abnormal data correction.

[0142] Next, the optimization control module uses the pre-processed data without abnormalities to perform calculations, sends new operating instructions to the control center of each converter station through Ethernet, and records data, analyzes the system operating status and handles abnormalities. At the same time, the optimization control module can also receive external instructions through the human-computer interaction module and perform visual display.

[0143] Finally, each converter station control center determines locally whether to execute the control instructions issued by the energy management and control system, and forms a closed-loop operation of the overall system after making corresponding adjustments.

[0144] 2. Core algorithm function module of urban rail intelligent energy management system

[0145] The system architecture of the core control module is as follows Figure 2As shown in the figure. The information of the power supply system and the vehicle provided by the real-time monitoring system passes through the data processing module as input to activate the core control module. It should be pointed out that the data processing module and the core control module use dual-threaded parallel processing, and each module has multiple sub-threads with different tasks to improve the overall rate of the algorithm.

[0146] The control module first passes through the static model fitting, and then combines the data information to comprehensively judge whether the system is in normal operation. If the system is in a fault state, different control strategies will be immediately adopted according to the fault level. When the system fault is not too serious, part of the power supply energy can be sacrificed, and the rest can still maintain economic optimization operation. When the system fault is serious, the control authority will be delegated to the converter station to ensure the stable operation of the system, thus giving up economy. The fault judgment time is within milliseconds, mainly relying on the automatic judgment of the system, and the control priority is the highest. This instruction can also be manually set by the system operator according to the real-time working conditions.

[0147] If it is in normal working state, the internal control algorithm will calculate the current optimization decision and send it to each converter station. The converter station adjusts the voltage at the station end according to the current strategy. The action cycle is more than one second. When the system determines that the fault has been cleared, the control platform automatically changes the mode from the fault state to the normal energy-saving working state. If the system fails again at this time, it will continue to enter the fault working mode. If the system fault occurs repeatedly several times, the command to switch to the normal working mode can only be manually issued by the system operator based on the actual situation, so the recovery command has the lowest priority.

[0148] In particular, there are some special working conditions when the system is in normal working state. For example, the problem of circulating current in the station caused by low voltage during energy feeding. At this time, the bottom control system will autonomously take over the control authority to suppress the circulating current, thereby ignoring the instructions issued by the top control module. When the circulating current condition is not triggered, the control authority will be automatically exchanged and the upper layer instructions will be obeyed. Therefore, the action cycle of this module is within milliseconds, and its priority is higher than that of the energy-saving module.

[0149] 3. Data verification method

[0150] The data processing module obtains data from power monitoring and vehicle-to-ground wireless systems every 0.5 seconds. Due to the time transmission deviation between different data sources, the timestamp of each data packet needs to be aligned. First, the vehicle data with a time difference of less than 300 milliseconds is considered to be the same time section data. Then, the vehicle data and bidirectional conversion data with a time difference of less than 30 milliseconds are considered to be the same time section data. Finally, considering that the train is the only load in the traction network, it is judged whether there is a serious data loss problem based on the total current of the traction network power supply and the total current of the vehicle power consumption. If the difference between the two is less than 500A, it is considered that the entire section data packet is complete and can be used for algorithm calculation. Otherwise, the entire package of data is discarded.

[0151] Since the collected data has measurement errors, data missing, ripple interference and other problems, data verification processing is required. This solution stores the data stack in a separate thread with a storage cycle of 0.5 seconds, and some individual data points may be lost during the process. Then, a smoothing spline fitting is performed on the entire 10-second data in another thread. For example, Figure 3 A schematic diagram of the voltage sampling value and the fitting curve of a substation is given in Figure 1. According to the obtained fitting curve function, the missing data in the storage process can be compensated, and the influence of ripple and measurement error can be filtered out.

[0152] It should be pointed out that there are many existing curve fitting methods. This scheme selects the Smoothing Spline method with better filtering characteristics for rail transit measurement systems.

[0153] 4. Static model setting method for rail transit power supply system

[0154] After completing the curve fitting of all the sampling data of the system, the system uses the latest fitting data at the current moment to perform static model tuning in a separate thread.

[0155] First consider the DC side static model.

[0156] When modeling a traction substation, it is usually equivalent to a Thevenin or Norton equivalent circuit in which the voltage source and the internal resistance change with the load. Then, according to the actual load of the traction substation, the working state of the rectifier unit is determined, that is, the conduction commutation mode of the rectifier bridge arm is determined, so as to calculate the equivalent internal resistance and ideal voltage source in the Thevenin equivalent circuit.

[0157] When modeling vehicles, because within the normal fluctuation range of the traction network voltage, the power of different vehicles at any same position or at the same time is independent of the network voltage, but the current flowing through the vehicle is directly affected by the actual network voltage. Therefore, the vehicle should be modeled as a power source, that is, the power obtained from the contact network by different vehicles at a certain static time section is fixed.

[0158] Therefore, the equivalent model of the traction station and the vehicle is as follows: Figure 4 shown.

[0159] When the current flowing into the drainage network or metal conductor is ignored, the current on the traction network is equal to the return current of the rail. Therefore, after the rail resistance is transferred to the traction power grid, the voltage difference between the traction network and the rail has a form similar to the traditional node potential equation. At the same time, at the traction substation, the up and down contact networks and the up and down running tracks are connected in parallel. Taking the Phase II of Jinan Metro Line 3 as an example, the line includes 5 traction substations including Chuanliu Station, Xiangyang Station, Lingang Station, Airport South Station and Yaoqiang Airport Station, and the following equations can be obtained: Figure 5The equivalent circuit of the DC side traction network of rail transit is shown.

[0160] based on Figure 5 , the node equation of the entire DC side traction network can be expressed as:

[0161] YU=I (1);

[0162] Since the vehicle is modeled as a power source, the admittance matrix of the entire system is nonlinear, and the solution of the node voltage requires multiple iterations. The specific solution process is as follows: Figure 6 shown.

[0163] For the AC side, the algorithm will perform power flow calculation by interfacing with the traction substation and adopting an AC / DC hybrid iterative method.

[0164] The equivalent circuit at the interface between the distributed AC distribution network and the DC is as follows: Figure 7 shown.

[0165] Where P s It is a 110kV main transformer and is treated as a relaxation node or a balance node in the AC power flow. Each converter station substation is modeled as a PQ node. It should be pointed out that due to the introduction of the bidirectional energy feeding device, the PQ node can be regarded as a four-quadrant node. Therefore, the mathematical model of the AC side distribution network is:

[0166]

[0167] Formula (2) is a classic AC power flow model, which is a nonlinear inequality and can be solved by the Newton-Raphson iterative method.

[0168] Usually, the traction load has little effect on the AC side voltage. Therefore, when calculating the DC side power flow, the AC voltage on the AC side can be considered unchanged. After the DC side power flow converges, the DC side is converted to PQ node decoupling through the AC / DC interface model of the traction substation, and then the active power, reactive power and current flowing into the substation on the AC side are calculated. The mathematical model of the interface between the AC side and the DC side is:

[0169]

[0170] Where φ is the power factor angle, T i is the traction transformer ratio, I dc The DC port of the traction substation supplies current. It is necessary to refer to: the tram load is only a small part of the load in the city network, and most of them adopt a distributed power supply method. If the AC side power flow is to be strictly calculated, other loads of the city network should be comprehensively considered. The traction substation can be regarded as a PQ node for power flow calculation.

[0171] 5. Optimization control algorithm for rail transit power supply system

[0172] The optimization control of the rail transit power supply system can be regarded as a nonlinear dynamic programming problem as a whole. The static model obtained in the previous section is an equality constraint, and the voltage and power limits are inequality constraints. By adjusting the substation voltage, the following optimization objectives are achieved:

[0173] i) Maximize the utilization of regenerative energy

[0174] Maximizing the utilization of renewable energy is mainly achieved by adjusting the voltage of the converter station, reducing the energy lost in the braking resistor during the vehicle braking phase, and allowing the regenerated energy to be absorbed by other vehicles in the traction network as much as possible.

[0175] A more detailed vehicle equivalent circuit is shown in Figure 8 As shown, r_br is the braking resistor, C f is the supporting capacitance, r f and L f is the equivalent impedance from the vehicle to the traction network. When the vehicle is in the deceleration stage, the current flows from the vehicle to the traction network, the capacitor is charged and the network voltage is raised. If the network voltage is too high, the switch will be closed, and the extra energy will be absorbed by the braking resistor and dissipated as heat. Therefore, it cannot be converted into renewable energy that can be reused. Although the vehicle braking time is relatively short, the instantaneous power is large due to the high DC side voltage and small braking resistor, so the regenerative energy that can be maximized is very considerable.

[0176] Assume that the vehicle power during deceleration is P i >0, the voltage threshold for closing the brake resistor is V br , considering r f and L f The impact of the vehicle on the traction grid side voltage threshold can be set to V br -δ br Among them, δ br It represents the voltage drop on the equivalent resistance of the vehicle network, which can be adjusted dynamically according to the current value. Based on this, the calculation formula of the braking resistance loss is:

[0177]

[0178] ii) Minimize power line transmission losses;

[0179] The main goal of minimizing power line transmission losses is to reduce the loss of power during transmission by controlling the voltage of the converter station. Line loss is the difference between power generation and power consumption. The line loss on the DC side and the AC side can be considered separately. The calculation formula is:

[0180]

[0181] in, is the total AC and DC power generation, is the sum of AC and DC loads. Although the line power loss is relatively small, due to its long-term existence, the cumulative energy consumption reduction is also considerable.

[0182] iii) Suppressing the intra-station circulation;

[0183] The schematic diagram of the circulation in the station is as follows Fig. 9 As shown, its triggering mechanism is similar to the braking resistor triggering condition, but it is mainly for substations. When the bidirectional energy feeding device is in the state of feeding power to the grid and the voltage is lower than the open-circuit voltage of the rectifier, the energy will be dissipated by the circulating current between the bidirectional energy feeding device and the rectifier.

[0184] When the open circuit voltage of the rectifier is U oc When , the calculated power of the circulating current in the station is

[0185]

[0186] It should be pointed out that the open circuit voltage U oc It is related to the AC side grid voltage, so it can also be dynamically updated according to the system operation status. In addition, although the energy management system will suppress the circulating current found at the current moment, the circulating current in the station will still appear in the gap between the two optimization instructions. Therefore, the management system regards the circulating current in the station as a special operating condition. That is, once the circulating current in the station occurs, the converter station will take control measures to suppress the circulating current in the station. At this time, the bottom control system will autonomously take over the control authority to suppress the circulating current, thereby ignoring the instructions issued by the top control module.

[0187] iv) Large double-sided voltage regulation control

[0188] When the energy management system detects that a rectifier is down, the adjacent node converter station will provide the corresponding voltage support and appropriately ignore the optimization operation mode. Assume that the rated voltage of site i is At this moment, the measured voltage is u i , then the voltage setting value of the adjacent node j, h converter station is adjusted to

[0189]

[0190] Among them, K j and K h is the droop ratio, which is determined by the capacity of stations j and h and the distance to station i.

[0191] It should be pointed out that objectives (i)-(iii) are all optimization objectives, and objective (iv) is a control measure. Therefore, the overall nonlinear dynamic programming modeling of the rail transit power supply system can be expressed as:

[0192]

[0193] Equation (8) is a typical nonlinear dynamic programming problem. The proposed solution uses the sequential quadratic programming (SQP) method to solve it according to the system characteristics. It should be noted that the tram power must be constant when solving the problem.

[0194] After solving equation (8), the optimal voltage setting value of each substation can be obtained: The set value is then used to consider whether the target voltage (iv) needs to be provided. If not, it is set as the target voltage value after fine-tuning according to the rectification droop characteristics according to formula (9);

[0195]

[0196] Where K_d is the droop slope, I_i^* is the rated current of device i, and I_i is the actual measured current.

[0197] Finally, the closed-loop optimization control of the rail transit power supply system is completed.

[0198] The intelligent control method of the rail transit traction power supply system of the present invention has the following beneficial effects:

[0199] 1. The patented method proposes to use multi-threaded parallel processing to achieve real-time optimization control of the rail transit power supply system, and proposes a closed-loop industrial application solution for data from entry to exit; a multi-threaded parallel processing method is used to ensure the dynamic optimization operation of the urban rail energy system.

[0200] 2. By analyzing the data characteristics of the rail transit power supply system, a more effective data preprocessing method is proposed; the patented algorithm proposes a data preprocessing method to perform timestamp alignment, data verification, missing value compensation, and bad data filtering on the rough sampling data.

[0201] 3. The patented method establishes a static power flow model for the AC and DC sides of the rail transit power supply system, simplifies the complex and abstract high-dimensional nonlinear physical system into theoretically solvable mathematical equations, and presents the complex power supply system using simplified mathematical expressions.

[0202] 4. Based on the derived system static model, the patented algorithm further established a nonlinear dynamic programming optimization control model, and considered multiple control objectives including maximizing the utilization of renewable energy, minimizing the power line transmission loss, and suppressing the circulation within the station; it took into account multiple optimization objectives, while achieving substantial energy conservation and emission reduction, improving the power quality and ensuring a more stable operation of the system.

[0203] 5. The patented algorithm proposes a more general methodology that can flexibly integrate more future optimization goals and control resources.

[0204] 6. The patented algorithm has been verified through integrated real-time simulation and will be applied in the Phase II of Jinan Metro Line 3 to form an engineering product.

[0205] The above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention is described in detail with reference to the preferred embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the present invention, which should be included in the scope of the technical solution for protection of the present invention.

Claims

1. An intelligent control method for rail transit traction power supply system, The following steps are involved: S10, data collection: real-time collection of power monitoring system data, signal system data and train operation data; S20, data processing: pre-processing the collected real-time data, including delay compensation and abnormal data correction; S30, optimization control: based on the compensated and corrected data, use intelligent algorithms to calculate and determine the power distribution plan of the DC network; S40, issuing instructions: converting the optimized power allocation scheme into control instructions, and sending them to the control center of each converter station via Ethernet; S50, system adjustment: each converter station control center adjusts the station terminal voltage according to the received control command to form a system closed-loop operation; Wherein, in step S20, delay compensation includes: synchronizing the data by comparing the timestamps of different data sources so as to align them under the same time reference; Abnormal data correction includes: filtering and correcting abnormal data points that exceed the normal range by setting a threshold range.

2. According to claim 1, the intelligent control method for the rail transit traction power supply system, It is characterized in that The optimization control steps include: Using nonlinear dynamic programming algorithm, combined with the physical constraints and objective function of the system, the power distribution of the DC grid is optimized; Among them, the nonlinear dynamic programming algorithm for optimal control is described by the following formula: DC side power flow calculation: YU=I (1); Where Y is the node admittance matrix, U is the node voltage vector, and I is the injected current vector; The AC side power flow calculation is: Among them, P s is the power generation, V s is the voltage, G ij and B ij are the real and imaginary parts of the admittance matrix, θ ij is the voltage phase angle difference; The mathematical model of the interface between the AC side and the DC side is: φ is the power factor angle, T i is the traction transformer ratio, I dc The DC port of the traction substation supplies current.

3. According to claim 2, the intelligent control method for the rail transit traction power supply system, It is characterized in that In step S30, the control algorithm is optimized to achieve the following goals: Maximize the utilization rate of regenerative energy, including: adjusting the voltage of the converter station and reducing the energy lost in the braking resistor during the vehicle braking phase; The calculation formula for brake resistance loss is: Assume that the vehicle power during deceleration is P i >0, the voltage threshold for closing the brake resistor is V br , considering r f and L f The impact of the vehicle on the traction grid side voltage threshold can be set to V br -δ br ; Among them, δ br Represents the voltage drop on the equivalent resistance of the vehicle network; P br (u i ) is the braking resistor loss, u i is the voltage, V ub is the voltage threshold, δ br is the pressure drop, P i is power, r br is the braking resistor; Minimize the power line transmission loss, including: controlling the voltage of the converter station to reduce the loss of power during transmission; line loss is the difference between power generation and power consumption, calculated by the following formula: Among them, P loss (u i ) is the line loss, is the total AC and DC power generation, is the total AC and DC loads; Suppressing the circulating current in the station, including: when the bidirectional energy feeding device is in the state of feeding power to the grid and the voltage is lower than the open circuit voltage of the rectifier, the energy will be dissipated by the circulating current between the bidirectional energy feeding device and the rectifier. When the open circuit voltage of the rectifier is U oc When , the calculated power of the circulating current in the station is: And, large bilateral voltage stabilization control, including: when a rectifier is detected to be down, the adjacent node converter station will provide voltage support and ignore the optimization operation mode; assuming that the rated voltage of site i is At this moment, the measured voltage is u i , then the voltage setting value of the adjacent node j, h converter station is adjusted to: Among them, K j and K h is the droop ratio, which is determined by the capacity of stations j and h and the distance to station i.

4. According to claim 3, the intelligent control method for the rail transit traction power supply system, It is characterized in that The overall modeling of the nonlinear dynamic programming of the rail transit power supply system is expressed as: Among them, the sequential quadratic programming (SQP) method is used to solve the above equation according to the system characteristics. When solving, the tram power is kept constant, and the optimal voltage setting value of each substation can be obtained.

5. According to claim 4, the intelligent control method for rail transit traction power supply system, It is characterized in that Sure Whether it is necessary to make adjustments according to formula (7); if no adjustment is required, then fine-tune according to formula (9) based on the rectification droop characteristics and set it as the target voltage value; In the formula, K d is the droop slope, is the rated current of device i, I i It is the actual measured current.

6. The intelligent control method for rail transit traction power supply system according to claim 5, It is characterized in that The steps of issuing instructions include: The optimized power allocation scheme is converted into specific control instructions and sent to the control center of each converter station via Ethernet for execution.

7. The intelligent control method for rail transit traction power supply system according to claim 6, It is characterized in that The system adjustment steps include: The control center of each converter station receives and executes control instructions to adjust the voltage at the station end.

Citation Information

Cited By

  • Droop control method and control equipment of traction power supply system and traction power supply system

    CN122338829A