Railway power regulator optimal compensation current command value calculation method

Abstract

The application discloses a kind of railway power regulator optimal compensation current instruction value calculation method, the scheme is according to RPC compensation power minimum as optimization target, with three-phase voltage unbalance degree, power factor, negative sequence current and so on Index parameter as inequality constraint, with converter compensation capacity, power conservation and so on As equality constraint, a series of different optimization mathematical models can be established from this, and then based on optimization algorithm real-time solution is carried out.Makes under different working conditions, the current reference value of railway power regulator is more convergent accurate, to realize better energy management control effect of photovoltaic power generation system of traction.

Description

Technical Field

[0001] This invention belongs to the field of electrified railway technology, and specifically relates to a method for calculating the optimal compensation current command value of a railway power regulator. Background Technology

[0002] High demand capacity and investment costs are technical bottlenecks restricting the large-scale application of railway power regulators (RPCs). Therefore, researching how to obtain the optimal compensation effect within a limited capacity range has significant practical engineering implications. Currently, RPC capacity optimization methods mainly include the following two types: 1. Capacity optimization based on optimized compensation methods; 2. Capacity optimization based on electrical structure improvements. Since most research on improving the electrical structure involves connecting low-cost passive branches in series on the connection branch between the converter and the transformer, including thyristor-controlled reactors (TCRs), thyristor-switched capacitors (TSCs), and filter capacitors (FCs), this design exacerbates the complexity of active and passive compensation capacity control. Therefore, current mainstream research focuses on the first approach.

[0003] Existing research often uses minimizing RPC compensation power as the optimization objective, employing parameters such as three-phase voltage imbalance and power factor as inequality constraints, and converter capacity and power conservation as equality constraints. This allows for the establishment of a series of different optimization mathematical models, which are then solved in real-time using optimization algorithms. However, such research suffers from the following problems: When determining the optimal compensation power reference value for converters, early analytical and enumeration methods struggle to meet the dual requirements of search speed and accuracy. Therefore, intelligent algorithms with strong optimization capabilities and fast convergence speeds are needed. Some studies utilize Sequential Quadratic Programming (SQP) as an effective method for solving nonlinear constraint problems, possessing strong local search capabilities, but its global search capability is weak and it is sensitive to initial conditions; using it alone may not find the true optimal solution. Other studies, to adapt to multidimensional strongly coupled time-varying control in optimization compensation models, have adopted robust and convenient fuzzy control, but this control design lacks systematicity and cannot define control objectives. Still other studies have used differential evolution algorithms to solve for the optimal compensation power reference value, but like other population-based algorithms, it suffers from premature convergence and insufficient local search capabilities. Similarly, the particle swarm optimization algorithm also suffers from premature convergence and insufficient local search capability. However, due to its relatively fast speed of approximating the optimal solution, it can effectively optimize system parameters and is therefore widely used in the calculation of reference values ​​in optimization compensation models. Therefore, it is necessary to design a method for calculating the optimal compensation current command value of railway power regulators to solve the above problems. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a method for calculating the optimal compensation current command value of a railway power regulator. This method aims to improve the commonly used PSO algorithm, solve the problem that the existing technology cannot take into account both global and local search capabilities, and that the PSO algorithm has the problem of premature convergence. It has the characteristics of being applicable to AC non-phase traction power supply systems, and while solving the optimal compensation power of the RPC, it can also use auxiliary parameters to further accelerate the real-time compensation of the RPC.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for calculating the optimal compensation current command value of a railway power regulator includes the following steps: S1. Construct a compensation power optimization model for the railway power regulator in the traction power supply system; S2, the system's main controller first calculates the traction load power and the target output value and unused electricity of the photovoltaic power generation unit based on the detected real-time voltage and current data of the traction side and photovoltaic side by the system's integrated energy management module. S3 utilizes a back-to-back converter optimization compensation model, introducing dynamically adjusted inertia weights and learning factors into the particle swarm optimization algorithm, and calculating the results using the new particle swarm optimization algorithm. Harmony Compensation power reference value of phase converter ( , , Then, using the compensation current calculation formula, the reference value of the compensation current is obtained. and ;in, for Phase converter compensation active power reference value for Phase converter reactive power compensation reference value, for Phase converter compensation active power reference value for Reference value for reactive power compensation by phase converter.

[0006] Preferably, in step S1, the compensation power optimization model for the railway power regulator is as follows: ; ; In the formula, Indicates: The maximum negative sequence current corresponding to the three-phase voltage imbalance limit on the system side is used as the reference value for normalization. The instantaneous power factor on the system side after photovoltaic integration. For negative sequence current, min PThis represents the minimum power compensation for the RPC. for Phase converter power compensation, for Phase converter compensation power; ; ; In the formula, and They are respectively Harmony Maximum capacity of phase converter and The active power compensated by the converter. and The reactive power compensated for by the converter.

[0007] Preferably, in step S3, the compensation model of the back-to-back converter is optimized, and the compensation power reference value of the two-phase converter is calculated by an improved particle swarm optimization algorithm. The specific method for solving the compensation power reference value is as follows: Step 301: Import real-time data from the traction side and the photovoltaic side, and obtain the corresponding traction load power, as well as the target output value and unused electricity of the photovoltaic power generation unit from the system integrated energy management module; Step 302, Initialize the particle swarm: Set the population size Number of iterations Randomly initialize particle positions ,speed Individual optimal position and the optimal position of the group ; Step 303: Introduce dynamically adjusted inertia weights, individual learning factors, and swarm learning factors into the particle swarm algorithm. Step 304: Calculate the particle fitness based on the optimization objectives of maximizing the power factor, minimizing the negative sequence current, and minimizing the compensation power of the railway power regulator. ; Step 305: Update the current optimal position of the particle based on particle fitness. and the optimal position of the group ; Step 306: Update the velocity of each particle according to the iterative formula of the particle swarm optimization algorithm, the compensation power constraint, and the three-phase voltage imbalance index constraint. and location The iterative formula for the particle swarm optimization algorithm is: ; In the formula, For the first Sub-iteration particles The first of the flight velocity vectors Dimensional components; For the first Sub-iteration particles The position vector of the first Dimensional components; and These are individual and group learning factors, respectively. and A random number that takes values ​​in the range [0,1]. Inertial weights; Step 307: When the maximum number of iterations is reached or the global optimal position is reached within the minimum range, the iterative calculation terminates, and the optimal compensation power reference value is obtained accordingly. Step 308: The compensation current command value of the two-phase converter is obtained through the compensation current calculation formula.

[0008] Preferably, in step S3, dynamically adjusted inertia weights and learning factors are introduced into the particle swarm optimization algorithm. The dynamic inertia weights and learning factors are as follows: ; Dynamic weights: 0.9~0.4; Self-awareness factor: 2 ~ 2.5; Social factor 2 ~ 2.5.

[0009] The beneficial effects of this invention are as follows: This invention proposes a method for calculating the optimal compensation current command value of a railway power regulator. An optimized compensation model is established with the optimization objectives of minimizing RPC compensation power, reducing negative sequence current, increasing power factor, and three-phase imbalance index. The improved PSO algorithm is used to calculate the reference compensation power value of the two-phase converters in the railway power regulator, and then the reference compensation current value is obtained using the compensation current calculation formula. In the initial iteration phase of the improved algorithm, the inertia weight is excessively large to encourage global search, while the learning factor is small to avoid prematurely falling into local optima. As the number of iterations increases, the inertia weight decreases to encourage local search, while the learning factor increases to encourage particles to search towards individual and global optima, thus balancing the needs of local and global searches. This method can more accurately and in real-time reflect the compensation energy in the railway power regulator, achieving comprehensive management of power quality issues such as negative sequence and reactive power, reducing the high demand capacity and investment cost of the RPC, and improving the long-term operating capability of the RPC, showing good prospects for industrial applications. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of the calculation process for the compensation power reference value of the back-to-back converter in this invention. Figure 2 This is a schematic diagram of the photovoltaic power supply system for traction in this invention. Detailed Implementation

[0011] Example 1: like Figure 1 As shown, a method for calculating the optimal compensation current command value of a railway power regulator includes the following steps: S1. Construct a compensation power optimization model for the railway power regulator in the traction power supply system; S2, the system's main controller first calculates the traction load power and the target output value and unused electricity of the photovoltaic power generation unit based on the detected real-time voltage and current data of the traction side and photovoltaic side by the system's integrated energy management module. S3 utilizes a back-to-back converter optimization compensation model, introducing dynamically adjusted inertia weights and learning factors into the particle swarm optimization algorithm, and calculating the results using the new particle swarm optimization algorithm. Harmony Compensation power reference value of phase converter ( , , Then, using the compensation current calculation formula, the reference value of the compensation current is obtained. and ;in, for Phase converter compensation active power reference value for Phase converter reactive power compensation reference value, for Phase converter compensation active power reference value for Reference value for reactive power compensation by phase converter.

[0012] Preferably, in step S1, the compensation power optimization model for the railway power regulator is as follows: ; ; In the formula, Indicates: The maximum negative sequence current corresponding to the three-phase voltage imbalance limit on the system side is used as the reference value for normalization. The instantaneous power factor on the system side after photovoltaic integration. For negative sequence current, min P This represents the minimum power compensation for the RPC. for Phase converter power compensation, for Phase converter compensation power; ; ; In the formula, and They are respectively Harmony Maximum capacity of phase converter and The active power compensated by the converter. and The reactive power compensated for by the converter.

[0013] Preferably, in step S3, the compensation model of the back-to-back converter is optimized, and the compensation power reference value of the two-phase converter is calculated by an improved particle swarm optimization algorithm. The specific method for solving the compensation power reference value is as follows: Step 301: Import real-time data from the traction side and the photovoltaic side, and obtain the corresponding traction load power, as well as the target output value and unused electricity of the photovoltaic power generation unit from the system integrated energy management module; Step 302, Initialize the particle swarm: Set the population size Number of iterations Randomly initialize particle positions ,speed Individual optimal position and the optimal position of the group ; Step 303: Introduce dynamically adjusted inertia weights, individual learning factors, and swarm learning factors into the particle swarm algorithm. Step 304: Calculate the particle fitness based on the optimization objectives of maximizing the power factor, minimizing the negative sequence current, and minimizing the compensation power of the railway power regulator. ; Step 305: Update the current optimal position of the particle based on particle fitness. and the optimal position of the group ; Step 306: Update the velocity of each particle according to the iterative formula of the particle swarm optimization algorithm, the compensation power constraint, and the three-phase voltage imbalance index constraint. and location The iterative formula for the particle swarm optimization algorithm is: ; In the formula, For the first Sub-iteration particles The first of the flight velocity vectors Dimensional components; For the first Sub-iteration particles The position vector of the first Dimensional components; and These are individual and group learning factors, respectively. and A random number that takes values ​​in the range [0,1]. Inertial weights; Step 307: When the maximum number of iterations is reached or the global optimal position is reached within the minimum range, the iterative calculation terminates, and the optimal compensation power reference value is obtained accordingly. Step 308: The compensation current command value of the two-phase converter is obtained through the compensation current calculation formula.

[0014] Preferably, in step S3, dynamically adjusted inertia weights and learning factors are introduced into the particle swarm optimization algorithm. The dynamic inertia weights and learning factors are as follows: ; Dynamic weights: 0.9~0.4; Self-awareness factor: 2 ~ 2.5; Social factor 2 ~ 2.5.

[0015] Example 2: like Figure 2 As shown, the traction photovoltaic power supply system includes a railway power regulator. The regulator adopts a back-to-back structure, with its AC input connected to the 220 kV double feeder of the traction power supply system via a step-down transformer. Its common capacitor's DC end is connected to the photovoltaic power generation device and the surplus energy storage device (denoted as...). ).

[0016] The power supplied by the power system side is respectively , , Mutually / The phase traction load power are respectively , ; Mutually / The active power output of the phase converter is respectively , The total electrical energy generated by the photovoltaic power generation unit is External output power is Unutilized power is ,and .

[0017] The system will fully consider the operating conditions of the photovoltaic and energy storage system, as well as train traction, braking, no-load, or coasting conditions. Energy management units will be set up for these conditions, with the management unit based on the target output value of the photovoltaic power generation unit. and remaining stored energy To examine the parameters.

[0018] Then, based on the detected real-time voltage and current data of the traction side / photovoltaic side, the energy management unit calculates the parameters to be examined.

[0019] Next, an optimized compensation model is established. The primary task of optimized compensation is to analyze the relationship between power quality index parameters (three-phase voltage imbalance, power factor, negative sequence current) and the given value of compensation power, and to establish an optimized compensation mathematical model.

[0020] 1. Three-phase voltage imbalance constraint: According to the requirements of "Power Quality Three-phase Voltage Imbalance", the three-phase voltage imbalance at the point of common coupling (PCC) is... It can be obtained from the negative sequence current value on the grid side. Indirect reaction:

[0021] In the formula, The rated line voltage of the three-phase power grid. This refers to the short-circuit capacity of the power supply.

[0022] 2. Negative sequence current, power factor constraint: ; 3. Compensation power constraint: ; ; ; The optimized compensation model for back-to-back converters is used, and the improved PSO algorithm is employed to calculate... Mutually / Compensation power reference value of phase converter ( , , ).

[0023] like Figure 1 As shown, the calculation process for the reference value of the compensation power of a back-to-back converter is as follows: 1. Improve the PSO algorithm calculation process by first importing real-time voltage and current data from the traction side / photovoltaic side, as well as power reference values ​​obtained from the power unit.

[0024] 2. Initialize the particle swarm, create a particle structure array, and randomly initialize particle positions. x Randomly initialize particle velocitiesv The optimal position of an individual particle is initialized to infinity, and the optimal position of an individual particle is initialized to its initial position.

[0025] 3. Calculate particle fitness based on three-phase voltage imbalance constraints, negative sequence current, power factor constraints, and compensation power constraints.

[0026] 4. Update the current particle's optimal position and the swarm's optimal position based on fitness. If the current position function value is less than the individual optimal position, update the individual optimal position. If the current position function value is less than the global optimal position, update the global optimal position.

[0027] 5. Next, update the particle velocity and position.

[0028] 6. When the maximum number of iterations is reached, or the global optimal position is found within the smallest range, exit the improved particle swarm algorithm iteration.

[0029] 7. Based on the obtained optimal power reference value, the optimal current reference value is obtained using the compensation current calculation formula.

[0030] The above embodiments are merely preferred technical solutions of this invention and should not be considered as limitations on this invention. The embodiments and features described in this application can be arbitrarily combined without conflict. The scope of protection of this invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of this invention.

Claims

1. A method for calculating the optimal compensation current command value of a railway power regulator, characterized in that: Includes the following steps: S1, Construct a compensation power optimization model for the railway power regulator in the traction power supply system; the compensation power optimization model is as follows: ； ； In the formula, This indicates that the maximum negative sequence current corresponding to the three-phase voltage imbalance limit on the system side is used as the reference value for normalization; PF is the instantaneous power factor on the system side after photovoltaic access; It is a negative sequence current; This is the minimum compensation power for the railway power regulator; for Phase converter power compensation, for Phase converter compensation power; ； ； In the formula, and They are respectively Harmony Maximum capacity of phase converter; and The active power compensated for by the converter; and Reactive power compensated for by the converter; S2, the system's main controller first calculates the traction load power and the target output value and unused electricity of the photovoltaic power generation unit based on the detected real-time voltage and current data of the traction side and photovoltaic side by the system's integrated energy management module. S3 utilizes a back-to-back converter optimization compensation model, introducing dynamically adjusted inertia weights and learning factors into the particle swarm optimization algorithm, and calculating the results through an improved particle swarm optimization algorithm. Harmony Compensation power reference value of phase converter ( , , , Then, using the compensation current calculation formula, the reference value of the compensation current of the converter is obtained. , ;in, for Phase converter compensation active power reference value for Phase converter reactive power compensation reference value, for Phase converter compensation active power reference value for Reference value for reactive power compensation by phase converter; The specific method for iteratively solving the power reference value using the particle swarm optimization algorithm in step S3 is as follows: Step 301: Import real-time data from the traction side and the photovoltaic side, and obtain the corresponding traction load power, as well as the target output value and unused electricity of the photovoltaic power generation unit from the system integrated energy management module; Step 302, Initialize the particle swarm: Set the population size Number of iterations Randomly initialize particle positions ,speed Individual optimal position and the optimal position of the group ; Step 303: Introduce dynamically adjusted inertia weights, individual learning factors, and swarm learning factors into the particle swarm algorithm. Step 304: Calculate the particle fitness based on the optimization objectives of maximizing the power factor, minimizing the negative sequence current, and minimizing the compensation power of the railway power regulator. ; Step 305: Update the current optimal position of the particle based on particle fitness. and the optimal position of the group ; Step 306: Update the velocity of each particle according to the iterative formula of the particle swarm optimization algorithm, the compensation power constraint, and the three-phase voltage imbalance index constraint. and location The iterative formula for the particle swarm optimization algorithm is: ； In the formula, For the first Sub-iteration particles The first of the flight velocity vectors Dimensional components; For the first Sub-iteration particles The position vector of the first Dimensional components; and These are individual and group learning factors, respectively. and A random number that takes values ​​in the range [0,1]. Inertial weights; Step 307: When the maximum number of iterations is reached or the global optimal position is reached within the minimum range, the iterative calculation terminates, and the optimal compensation power reference value is obtained accordingly. Step 308: The compensation current command value of the two-phase converter is obtained through the compensation current calculation formula.

2. The method for calculating the optimal compensation current command value of a railway power regulator according to claim 1, characterized in that: The particle swarm optimization algorithm introduces dynamically adjusted inertia weights and learning factors, which are as follows: ； Dynamic weights: 0.9~0.4; Self-awareness factor: 2 ~ 2.5; Social factor 2 ~ 2.5.

Images