Harmonic traceability method and device based on electrified railway and new energy inverter
By employing the Newton-Raphson method, harmonic power flow calculation method, Latin hypercube sampling, and deep belief network model, the dominant equipment type of harmonic sources in electrified railway and new energy inverters can be quickly identified, solving the problem of cumbersome acquisition process in existing technologies and improving efficiency and accuracy.
Patent Information
- Application Number
- CN202511704293.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-19
- Publication Date
- 2026-01-16
AI Technical Summary
The existing technology for obtaining the dominant equipment types for harmonic sources of electrified railways and new energy inverters is cumbersome, consumes a lot of human resources and time, and results in low efficiency.
The fundamental and harmonic admittance matrices of the power system are processed using the Newton-Raphson method and harmonic power flow calculation method. Combined with the Latin hypercube sampling strategy and deep belief network model, the harmonic distortion rate and voltage spectrum of each node are generated through the harmonic emission model of electrified railway and new energy inverter. The dominant equipment type of harmonic source is identified by the trained deep belief network model.
It can quickly identify the dominant device type of harmonic sources without human intervention, improving acquisition efficiency and accuracy of harmonic responsibility allocation.
Smart Images
Figure CN121350846A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of power quality analysis technology and artificial intelligence technology, and in particular to a harmonic source tracing method and device based on electrified railway and new energy inverters. Background Technology
[0002] Electrified railways and new energy inverters, as important components of the modern energy system, play a key role in promoting transportation electrification and clean energy transformation. However, during their operation, they inject harmonics into the power system. These harmonics propagate in the power system and affect the normal operation of various electrical equipment, leading to a decline in the performance of various electrical equipment. In order to locate the source of harmonics, it is necessary to obtain the dominant equipment type of the harmonic source.
[0003] However, the process of obtaining the dominant device type of existing harmonic sources is cumbersome, which is not conducive to improving the efficiency of obtaining the dominant device type of harmonic sources. The reason is that existing technologies mainly use manual methods to obtain the dominant device type of harmonic sources, which consumes a lot of human and time resources, increases the time required to obtain the dominant device type of harmonic sources, and is not conducive to improving the efficiency of obtaining the dominant device type of harmonic sources. Summary of the Invention
[0004] This application provides a harmonic source tracing method and apparatus based on electrified railway and new energy inverters to solve the technical problem that the process of obtaining the dominant equipment type of the existing harmonic source is cumbersome and not conducive to improving the efficiency of obtaining the dominant equipment type of the harmonic source.
[0005] In a first aspect, embodiments of this application provide a harmonic source tracing method based on electrified railways and new energy inverters, applied to electronic equipment, the harmonic source tracing method comprising: Acquire power system data and construct fundamental and harmonic admittance matrices based on the acquired data; The fundamental admittance matrix is processed using the Newton-Raphson method to obtain the effective voltage values of the fundamental components at each node in the power system. The harmonic power flow calculation method is used to process the harmonic admittance matrix to obtain the effective voltage values of the harmonic components at each node in the power system. Based on the effective voltage values of the harmonic components at each node, the effective voltage values of the fundamental components at each node, and the total harmonic distortion rate model, the total harmonic distortion rate of each node is generated. Nodes with harmonic distortion rates greater than the preset distortion rate are marked as harmonic exceedance points. When the number of harmonic exceedance points is a positive integer, the harmonic emission model of the electrified railway and the harmonic emission model of the new energy inverter are obtained from the simulation file. A Latin hypercube sampling strategy is adopted to sample multiple parameters of the harmonic sources of the power system, resulting in multiple sets of parameter combinations. These parameter combinations are then processed using a harmonic emission model of an electrified railway to obtain harmonic data for the electrified railway. Finally, the harmonic voltage spectrum of the electrified railway is obtained by processing the harmonic data using a three-phase power flow model of the electrified railway. Similarly, the harmonic data of the new energy inverter is obtained by processing multiple parameter combinations using a harmonic emission model of the new energy inverter. Finally, the harmonic voltage spectrum of the new energy inverter is obtained by processing the harmonic data of the new energy inverter using a three-phase power flow model of the new energy inverter. The harmonic voltage spectrum of the electrified railway is processed by a trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the electrified railway. Similarly, the harmonic voltage spectrum of the new energy inverter is processed by the same trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the new energy inverter. Based on these values and the generation model, the root mean square (RMS) values of the harmonic voltage spectrum of the electrified railway and the new energy inverter are generated. When the RMS value of the harmonic voltage spectrum of the electrified railway is greater than the harmonic threshold of the electrified railway and the RMS value of the harmonic voltage spectrum of the new energy inverter is not greater than the harmonic threshold of the new energy inverter, the dominant equipment type of the harmonic source is labeled as the electrified railway. Conversely, when the RMS value of the harmonic voltage spectrum of the electrified railway is not greater than the harmonic threshold of the electrified railway and the RMS value of the harmonic voltage spectrum of the new energy inverter is greater than the harmonic threshold of the new energy inverter, the dominant equipment type of the harmonic source is labeled as the new energy inverter.
[0006] In one possible implementation of the first aspect, the Newton-Raphson method is used to process the fundamental admittance matrix to obtain the effective voltage values of the fundamental components of each node in the power system; the harmonic power flow calculation method is used to process the harmonic admittance matrix to obtain the effective voltage values of the harmonic components of each node in the power system; and based on the effective voltage values of the harmonic components of each node, the effective voltage values of the fundamental components of each node, and the total harmonic distortion rate model, the total harmonic distortion rate of each node is generated, including: The fundamental admittance matrix is input into the power system simulation module. The simulation module uses the Newton-Raphson method to process the fundamental admittance matrix and obtain the effective voltage value of the fundamental component of each node in the power system. The harmonic admittance matrix is processed using the harmonic power flow calculation method to obtain the effective voltage values of the harmonic components at each node in the power system. Based on the effective voltage values of the harmonic components at each node, the effective voltage values of the fundamental components at each node, and the total harmonic distortion rate model, the total harmonic distortion rate of each node is generated.
[0007] In one possible implementation of the first aspect, the Latin hypercube sampling strategy is employed to sample multiple parameters of the harmonic sources of the power system, obtaining multiple sets of parameter combinations. These multiple sets of parameter combinations are then processed using a harmonic emission model of an electrified railway to obtain harmonic data for the electrified railway. The harmonic data are then processed using a three-phase power flow model of the electrified railway to obtain the harmonic voltage spectrum of the electrified railway. Finally, the multiple sets of parameter combinations are processed using a harmonic emission model of a new energy inverter to obtain the harmonic data of the new energy inverter. The harmonic data are then processed using a three-phase power flow model of the new energy inverter to obtain the harmonic voltage spectrum of the new energy inverter, including: Multiple parameters of the harmonic source are obtained from the configuration file. The Latin hypercube sampling strategy is used to sample multiple parameters of the harmonic source of the power system to obtain multiple sets of parameter combinations. Multiple sets of parameters are input into the harmonic emission model of the electrified railway through the first input interface. The harmonic emission model processes these parameters to obtain the harmonic data of the electrified railway. The harmonic data is then input into the three-phase power flow model of the electrified railway through the second input interface. The harmonic data is processed by the three-phase power flow model to obtain the harmonic voltage spectrum of the electrified railway. Multiple sets of parameters are input into the harmonic emission model of the new energy inverter through the third input interface. The harmonic data of the new energy inverter is then processed by the harmonic emission model of the new energy inverter to obtain the harmonic data of the new energy inverter. Finally, the harmonic data of the new energy inverter is input into the three-phase power flow model of the new energy inverter through the fourth input interface. The harmonic data is processed by the three-phase power flow model of the new energy inverter to obtain the harmonic voltage spectrum of the new energy inverter.
[0008] In one possible implementation of the first aspect, the new energy inverter includes a photovoltaic inverter and a wind power inverter.
[0009] In one possible implementation of the first aspect, the collected data includes topology data of the power network, parameters of power equipment, parameters of the grid connection interface of the electrified railway, and parameters of the grid connection interface of the new energy inverter. The parameters of the power equipment include transmission line parameters, transformer parameters, and generator parameters.
[0010] In one possible implementation of the first aspect, the harmonic emission model of an electrified railway is defined as follows: ; in, This indicates the current amplitude of the traction converter in an electrified railway. This refers to the grid-side voltage, which is the voltage provided by the power grid to electrified railways. It is the power factor. It is the power factor angle of the rectifier in an electrified railway; The traction power of electrified railways; The DC current output by the traction converter of an electrified railway to the traction motor; The switching frequency of the traction converter in an electrified railway; For industrial frequencies; Proportional parameters for the controller of electrified railways; Integral parameters of the controller for electrified railways; This refers to the operating mode coefficient for electrified railways. The value can be 1 or 0.8; A value of 1 indicates that the electrified railway is in traction mode; A value of 0.8 indicates that the electrified railway is in braking condition; The error signal for electrified railways is the difference between the reference value and the actual value of the current during the traction control process. The purpose of integrating the error signal of the electrified railway is to eliminate the steady-state error of the electrified railway and to ensure the stable operation of the traction converter.
[0011] In one possible implementation of the first aspect, the harmonic emission model of the new energy inverter is defined as follows: ; ; This indicates the harmonic current generated by the new energy inverter; This indicates the DC-side voltage of the new energy inverter; This represents the modulation amplitude parameter of the new energy inverter; This indicates the harmonic reactance of the new energy inverter, which is used to affect the flow of harmonic current; The modulation ratio of a new energy inverter is the ratio of its switching frequency to its fundamental frequency, and it is used to determine the modulation effect of the new energy inverter. This indicates a correction term for the controller of the new energy inverter, used to optimize the control performance of the new energy inverter; This indicates the switching frequency of the new energy inverter; This indicates the fundamental frequency of the new energy inverter; This represents the fundamental current output by the new energy inverter.
[0012] In one possible implementation of the first aspect, the generative model is defined as follows: ; For the first electrified railway The voltage spectrum amplitude of the subharmonic; For the first electrified railway The square of the voltage spectrum amplitude of the subharmonic; For the first new energy inverter The square of the voltage spectrum amplitude of the subharmonic; For the first new energy inverter The square of the voltage spectrum amplitude of the subharmonic; This represents the root mean square value of the harmonic voltage spectrum of an electrified railway. This represents the root mean square value of the harmonic voltage spectrum of the new energy inverter.
[0013] In one possible implementation of the first aspect, the total harmonic distortion rate model is defined as follows: ; This represents the total harmonic distortion rate of the i-th node. The higher the total harmonic distortion rate of the i-th node, the higher the degree of harmonic distortion of the i-th node; the lower the total harmonic distortion rate of the i-th node, the lower the degree of harmonic distortion of the i-th node. This represents the effective voltage value of the h-th harmonic component at the i-th node. This represents the square of the effective voltage value of all harmonic components at the i-th node; This represents the effective voltage value of the fundamental component at the i-th node.
[0014] Secondly, embodiments of this application provide a harmonic source tracing device based on electrified railway and new energy inverters, applied to electronic equipment, including: The first acquisition module is used to acquire the collected data of the power system and construct the fundamental wave admittance matrix and harmonic wave admittance matrix based on the collected data. The generation module is used to process the fundamental admittance matrix using the Newton-Raphson method to obtain the effective voltage value of the fundamental component of each node in the power system, process the harmonic admittance matrix using the harmonic power flow calculation method to obtain the effective voltage value of the harmonic component of each node in the power system, and generate the total harmonic distortion rate of each node based on the effective voltage value of the harmonic component of each node, the effective voltage value of the fundamental component of each node, and the total harmonic distortion rate model. The second acquisition module is used to mark nodes with harmonic distortion rates greater than preset distortion rates as harmonic exceedance points. When the number of harmonic exceedance points is a positive integer, the harmonic emission model of the electrified railway and the harmonic emission model of the new energy inverter are obtained from the simulation file. The sampling module is used to sample multiple parameters of the harmonic source of the power system using the Latin hypercube sampling strategy, obtaining multiple sets of parameter combinations. The multiple sets of parameter combinations are processed by the harmonic emission model of the electrified railway to obtain the harmonic data of the electrified railway. The harmonic data of the electrified railway are processed by the three-phase power flow model of the electrified railway to obtain the harmonic voltage spectrum of the electrified railway. The multiple sets of parameter combinations are processed by the harmonic emission model of the new energy inverter to obtain the harmonic data of the new energy inverter. The harmonic data of the new energy inverter are processed by the three-phase power flow model of the new energy inverter to obtain the harmonic voltage spectrum of the new energy inverter. The source tracing module processes the harmonic voltage spectrum of electrified railways using a trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the electrified railway. It then processes the harmonic voltage spectrum of the new energy inverter using the same trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the new energy inverter. Based on the voltage spectrum amplitudes of the electrified railway and new energy inverter harmonics, and the generation model, it generates the root mean square (RMS) values of the harmonic voltage spectrum of the electrified railway and the harmonic voltage spectrum of the new energy inverter. The root mean square (RMS) value of the harmonic voltage spectrum is used to identify the dominant equipment type of the harmonic source. When the RMS value of the harmonic voltage spectrum of the electrified railway is greater than the harmonic threshold of the electrified railway and the RMS value of the harmonic voltage spectrum of the new energy inverter is not greater than the harmonic threshold of the new energy inverter, the dominant equipment type of the harmonic source is identified as the new energy inverter.
[0015] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the harmonic source tracing method described in the first aspect above.
[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the harmonic source tracing method described in the first aspect above.
[0017] Fifthly, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to execute the harmonic source tracing method described in the first aspect.
[0018] The beneficial effects of this application's embodiments are twofold. Firstly, feature extraction is performed on the harmonic voltage spectrum of electrified railways to obtain a first feature vector, and feature extraction is performed on the harmonic voltage spectrum of new energy inverters to obtain a second feature vector. The first and second feature vectors are concatenated to obtain the harmonic feature vector of the power system. The harmonic feature vector is then processed by a trained deep belief network model to obtain the dominant equipment type of the harmonic source. Since manual acquisition is not required, the acquisition time of the dominant equipment type of the harmonic source is reduced, which is beneficial to improving the acquisition efficiency of the dominant equipment type of the harmonic source. Secondly, harmonic responsibility is assigned based on the dominant equipment type, which is beneficial to improving the accuracy of harmonic responsibility assignment. Attached Figure Description
[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1 This is an application scenario diagram of the harmonic source tracing method provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the harmonic source tracing method provided in an embodiment of this application; Figure 3 A flowchart illustrating the implementation of S202 provided in this application embodiment; Figure 4 A schematic block diagram of a harmonic source tracing device provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.
[0022] The harmonic source tracing method provided in this application can be applied to electronic devices such as mobile phones, tablets, wearable devices, vehicle-mounted devices, laptops, personal computers, and netbooks. This application does not impose any restrictions on the specific type of electronic device.
[0023] Please see Figure 1 , Figure 1 The application scenario diagram of the harmonic source tracing method provided in the embodiments of this application is described in detail below: Electronic devices access the data platform, acquire power system data from the platform, and construct the fundamental waveguide admittance matrix and harmonic waveguide admittance matrix based on the acquired data. In this embodiment, the electronic device obtains the power system's collected data from the data platform, reducing the data acquisition time and improving the efficiency of power system data acquisition.
[0024] Please see Figure 2 , Figure 2 This is a flowchart illustrating the harmonic source tracing method provided in this application embodiment, which can be applied to electronic devices.
[0025] like Figure 2 As shown, the harmonic source tracing method provided in this application includes the following steps, detailed below: S201, acquire the power system's data, and construct the fundamental waveguide admittance matrix and harmonic waveguide admittance matrix based on the acquired data; The collected data includes power network topology data, power equipment parameters, grid connection interface parameters of electrified railways, and grid connection interface parameters of new energy inverters. Power equipment parameters include transmission line parameters, transformer parameters, and generator parameters.
[0026] The new energy inverters include photovoltaic inverters and wind power inverters.
[0027] For example, before acquiring the power system's collected data and constructing the fundamental wave admittance matrix and harmonic wave admittance matrix based on the collected data, the harmonic source tracing method includes: Multiple harmonic samples are acquired, and these samples are combined into a sample set. The deep belief network model is trained using the sample set to obtain the trained deep belief network model. The trained deep belief network model is then saved to a model file.
[0028] S202 uses the Newton-Raphson method to process the fundamental admittance matrix, obtaining the effective voltage value of the fundamental component of each node in the power system. It uses the harmonic power flow calculation method to process the harmonic admittance matrix, obtaining the effective voltage value of the harmonic component of each node in the power system. Based on the effective voltage value of the harmonic component of each node, the effective voltage value of the fundamental component of each node, and the total harmonic distortion rate model, the total harmonic distortion rate of each node is generated. The total harmonic distortion rate model is defined as follows: ; This represents the total harmonic distortion rate of the i-th node. The higher the total harmonic distortion rate of the i-th node, the higher the degree of harmonic distortion of the i-th node; the lower the total harmonic distortion rate of the i-th node, the lower the degree of harmonic distortion of the i-th node. This represents the effective voltage value of the h-th harmonic component at the i-th node. This represents the square of the effective voltage value of all harmonic components at the i-th node; This represents the effective voltage value of the fundamental component at the i-th node.
[0029] S203, mark nodes with harmonic distortion rates greater than preset distortion rates as harmonic exceedance points. When the number of harmonic exceedance points is a positive integer, obtain the harmonic emission model of the electrified railway and the harmonic emission model of the new energy inverter from the simulation file. Among them, the harmonic emission model of electrified railways can simulate the harmonic emission level of electrified railways in different scenarios in advance.
[0030] Among them, the harmonic emission model of the new energy inverter can simulate the harmonic emission level of the new energy inverter in different scenarios in advance.
[0031] Among them, the harmonic emission model of electrified railways: The harmonic emission model of electrified railways can be established based on the switching function method or the known pulse width modulation spectrum characteristics. The harmonics of electrified railways mainly come from the traction converter.
[0032] The harmonic emission model for electrified railways is defined as follows: ; in, This indicates the current amplitude of the traction converter in an electrified railway. This refers to the grid-side voltage, which is the voltage provided by the power grid to electrified railways. It is the power factor. It is the power factor angle of the rectifier in an electrified railway; The traction power of electrified railways; The DC current output by the traction converter of an electrified railway to the traction motor; The switching frequency of the traction converter in an electrified railway; The industrial frequency is preferably 50Hz. Proportional parameters for the controller of electrified railways; Integral parameters of the controller for electrified railways; This refers to the operating mode coefficient for electrified railways. The value can be 1 or 0.8; A value of 1 indicates that the electrified railway is in traction mode; A value of 0.8 indicates that the electrified railway is in braking condition; The error signal for electrified railways is the difference between the reference value and the actual value of the current during the traction control process. The purpose of integrating the error signal of the electrified railway is to eliminate the steady-state error of the electrified railway and to ensure the stable operation of the traction converter.
[0033] The harmonic emission model of electrified railways can be used to conduct simulation analysis of electrified railways. During the design phase of electrified railways, parameter configurations can be optimized and control strategies can be adjusted to avoid excessive harmonics during operation and reduce interference to the power grid.
[0034] The harmonic emission model of a renewable energy inverter describes the harmonic current emitted by it under specific grid connection point voltage, output power, and control parameters. The harmonic emission model of a renewable energy inverter is typically established based on its control strategy and PWM modulation technology.
[0035] The harmonic emission model for new energy inverters is defined as follows: ; ; This indicates the harmonic current generated by the new energy inverter; This indicates the DC-side voltage of the new energy inverter; This represents the modulation amplitude parameter of the new energy inverter; This indicates the harmonic reactance of the new energy inverter, which is used to affect the flow of harmonic current; The modulation ratio of a new energy inverter is the ratio of its switching frequency to its fundamental frequency, and it is used to determine the modulation effect of the new energy inverter. This indicates a correction term for the controller of the new energy inverter, used to optimize the control performance of the new energy inverter; This indicates the switching frequency of the new energy inverter; This indicates the fundamental frequency of the new energy inverter; This represents the fundamental current output by the new energy inverter.
[0036] The harmonic emission model of the new energy inverter can be used to conduct simulation analysis of the new energy inverter. During the design stage of the new energy inverter, the parameter configuration can be optimized and the control strategy can be adjusted to avoid the generation of excessive harmonics during the operation of the new energy inverter and reduce the interference to the power grid.
[0037] S204 employs a Latin hypercube sampling strategy to sample multiple parameters of the harmonic sources in the power system, obtaining multiple parameter combinations. These combinations are then processed using a harmonic emission model of an electrified railway to obtain harmonic data for the electrified railway. This data is further processed using a three-phase power flow model of the electrified railway to obtain the harmonic voltage spectrum. Finally, multiple parameter combinations are processed using a harmonic emission model of a new energy inverter to obtain the harmonic data for the new energy inverter. This data is then processed using a three-phase power flow model of the new energy inverter to obtain the harmonic voltage spectrum for the new energy inverter. Latin Hypercube Sampling (LHS) is a multivariate parameter space sampling strategy based on a hierarchical approach. Its core logic lies in achieving efficient coverage of high-dimensional spaces through a combination of structured hierarchical processing and randomness. This method divides the domain of each input variable into multiple equally probable sub-intervals, and then randomly selects a sample point within each sub-interval, ensuring that each interval is sampled only once.
[0038] Among them, the three-phase power flow model is a tool used to analyze the voltage, current and power distribution in a three-phase AC power system. By constructing a set of nonlinear equations in a three-phase coordinate system, the three-phase power flow model can accurately simulate complex operating conditions such as three-phase imbalance, line parameter asymmetry and single-phase large load in the actual power grid.
[0039] The method employs a Latin hypercube sampling strategy to sample multiple parameters of the harmonic sources in the power system, obtaining multiple parameter combinations. These combinations are then processed using a harmonic emission model of an electrified railway to obtain harmonic data. The harmonic data is further processed using a three-phase power flow model of the electrified railway to obtain its harmonic voltage spectrum. Finally, a harmonic emission model of a new energy inverter is used to process these parameter combinations, yielding the new energy inverter's harmonic data. Finally, a three-phase power flow model of the new energy inverter is used to process the harmonic data, resulting in the new energy inverter's harmonic voltage spectrum. This process includes: Multiple parameters of the harmonic source are obtained from the configuration file. The Latin hypercube sampling strategy is used to sample multiple parameters of the harmonic source of the power system to obtain multiple sets of parameter combinations. Multiple sets of parameters are input into the harmonic emission model of the electrified railway through the first input interface. The harmonic emission model processes these parameters to obtain the harmonic data of the electrified railway. The harmonic data is then input into the three-phase power flow model of the electrified railway through the second input interface. The harmonic data is processed by the three-phase power flow model to obtain the harmonic voltage spectrum of the electrified railway. Multiple sets of parameters are input into the harmonic emission model of the new energy inverter through the third input interface. The harmonic data of the new energy inverter is then processed by the harmonic emission model of the new energy inverter to obtain the harmonic data of the new energy inverter. Finally, the harmonic data of the new energy inverter is input into the three-phase power flow model of the new energy inverter through the fourth input interface. The harmonic data is processed by the three-phase power flow model of the new energy inverter to obtain the harmonic voltage spectrum of the new energy inverter.
[0040] Harmonic amplitude refers to the strength of a harmonic signal, usually expressed as a specific value of voltage or current. Harmonic amplitude is directly related to the impact of harmonics on power grid equipment; for example, a higher amplitude is more likely to cause problems such as overheating and decreased accuracy of equipment.
[0041] Harmonic phase refers to the temporal relationship between harmonic signals and the fundamental signal of the power grid, measured by angular values. Harmonic phase affects the superposition effect of different harmonics in the power grid, thereby altering the actual operating state of the power grid, such as affecting power transmission efficiency.
[0042] Harmonic time series data, in particular, is data formed by continuously recording the changes in harmonic amplitude and phase at a certain monitoring point over a period of time. Harmonic time series data can show the changing patterns of harmonics over time, such as whether the harmonic amplitude increases or the phase fluctuates during a certain period of time. It is an important basis for finding the source of harmonics and judging the effectiveness of harmonic mitigation.
[0043] Using Latin hypercube sampling to process parameter combinations: Latin hypercube sampling is an efficient space-filling experimental design method for generating representative sample points in a multidimensional parameter space.
[0044] ①Sampling objects: The multiple parameters refer to the input variables in the above model. Latin hypercube sampling will combine the parameters for each parameter within its possible fluctuation range.
[0045] ② Model processing procedure: After obtaining N sets of parameter combinations, the processing procedure is direct and mechanical. Substitute the i-th set of parameter combinations into the established harmonic emission model, and output the harmonic current data corresponding to each harmonic under that set of operating conditions. Repeat the steps for all N sets of parameter combinations to obtain N sets of harmonic data corresponding to the uncertainty of the operating conditions.
[0046] 3. Establishment of Three-Phase Power Flow Model: A mathematical model capable of calculating harmonic power flow in a three-phase unbalanced power grid. Its core is the system's three-phase harmonic admittance matrix. The three-phase power flow model is constructed by assembling all components in the system according to the network topology. Mainstream power system analysis software has this function built-in, and can automatically construct this model based on the power grid's wiring diagram and component parameters.
[0047] The harmonic data obtained in the previous step is injected into the corresponding nodes in the network. Simultaneously, the fundamental voltage obtained from the fundamental power flow calculation in the first stage is used as the initial network condition for harmonic calculation. The harmonic power flow calculation is typically linearized. For the h-th harmonic, its network equation can be expressed as: ; It is the harmonic admittance matrix of the h-th harmonic; It is the inverse matrix of the h-th harmonic admittance matrix; It is the voltage vector of the h-th harmonic. It is the current vector of the h-th harmonic.
[0048] Solving the linear equations yields a complex voltage for each node and each harmonic order, with its magnitude representing the harmonic amplitude and its phase representing the harmonic phase. Furthermore, after obtaining the amplitudes and phases of all harmonic orders of interest, an inverse Fourier transform can be used to synthesize these frequency domain data into a time domain waveform, i.e., harmonic time series data.
[0049] S205: The harmonic voltage spectrum of the electrified railway is processed using a trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the electrified railway. Similarly, the harmonic voltage spectrum of the new energy inverter is processed using the same trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the new energy inverter. Based on the voltage spectrum amplitudes of the electrified railway harmonics, the new energy inverter harmonics, and the generation model, the root mean square value of the harmonic voltage spectrum of the electrified railway and the harmonic voltage of the new energy inverter are generated. The root mean square (RMS) value of the spectrum: When the RMS value of the harmonic voltage spectrum of the electrified railway is greater than the harmonic threshold of the electrified railway and the RMS value of the harmonic voltage spectrum of the new energy inverter is not greater than the harmonic threshold of the new energy inverter, the dominant equipment type of the harmonic source is marked as electrified railway. When the RMS value of the harmonic voltage spectrum of the electrified railway is not greater than the harmonic threshold of the electrified railway and the RMS value of the harmonic voltage spectrum of the new energy inverter is greater than the harmonic threshold of the new energy inverter, the dominant equipment type of the harmonic source is marked as new energy inverter.
[0050] When the root mean square value of the harmonic voltage spectrum of the electrified railway is greater than the harmonic threshold of the electrified railway, and the root mean square value of the harmonic voltage spectrum of the new energy inverter is not greater than the harmonic threshold of the new energy inverter, This indicates that the harmonic emissions of electrified railways have exceeded compliance standards, while the harmonic emissions of new energy inverters have not. The interference intensity of electrified railways on the power grid exceeds the standard, which may also lead to a decline in power quality at the point of common coupling and even trigger the risk of superimposed exceedance. Therefore, the dominant equipment type for identifying harmonic sources is electrified railways.
[0051] When the root mean square value of the harmonic voltage spectrum of the electrified railway is not greater than the harmonic threshold of the electrified railway, and the root mean square value of the harmonic voltage spectrum of the new energy inverter is greater than the harmonic threshold of the new energy inverter. This indicates that the harmonic emissions of electrified railways do not exceed compliance standards, while the harmonic emissions of renewable energy inverters do. The interference intensity of renewable energy inverters on the power grid exceeds standards and may also lead to a decline in power quality at the point of common coupling, or even trigger a cumulative risk of exceeding standards. Therefore, the dominant equipment type for identifying harmonic sources is renewable energy inverters. The generation model is defined as follows: ; For the first electrified railway The voltage spectrum amplitude of the subharmonic; For the first electrified railway The square of the voltage spectrum amplitude of the subharmonic; For the first new energy inverter The square of the voltage spectrum amplitude of the subharmonic; For the first new energy inverter The square of the voltage spectrum amplitude of the subharmonic; This represents the root mean square value of the harmonic voltage spectrum of an electrified railway. This represents the root mean square value of the harmonic voltage spectrum of the new energy inverter.
[0052] By identifying the dominant equipment type of harmonic sources, effective harmonic suppression strategies can be developed to reduce the impact of harmonics on voltage and current waveforms, making power quality closer to the ideal state, providing higher quality and more stable power supply for various electrical equipment, and reducing production interruptions and product quality declines caused by power quality problems.
[0053] The dominant equipment type of the harmonic source can determine harmonic responsibility. For example, if the dominant equipment type of the harmonic source is an electrified railway, then the electrified railway generates the main harmonics; if the dominant equipment type of the harmonic source is a new energy inverter, then the new energy inverter generates the main harmonics. This allows for the division of harmonic responsibility and the development of filtering schemes.
[0054] Following S205, the method also includes: Obtain the geographical location of the harmonic source and the probability statistics of the harmonic source, and integrate the dominant equipment type of the harmonic source, the geographical location of the harmonic source, and the probability statistics of the harmonic source into the harmonic source tracing result; The evaluation template is read from a preset file. The harmonic source tracing results are written into the evaluation template through the template engine's rendering function to obtain the power system evaluation report. A first window is created to display the evaluation report. A second window is created to display the harmonic mitigation scheme corresponding to the dominant equipment type.
[0055] For ease of explanation, the following example is provided: For example, the dominant equipment type of harmonic source is electrified railway, which is the dominant factor causing the fifth harmonic to exceed the standard.
[0056] For example, the geographical location of a harmonic source is the first area within the power grid's supply range.
[0057] For example, the probability statistics of harmonic sources are as follows: Under the current operating mode / scenario, the probability of the 17th harmonic resonance occurring is about 5%, and the voltage distortion rate at resonance may exceed 8%.
[0058] When the dominant equipment type is electrified railway, the harmonic mitigation scheme corresponding to electrified railway is displayed through the second window. When the dominant equipment type is new energy inverter, the harmonic mitigation scheme corresponding to new energy inverter is displayed through the second window. This effectively reduces the harmonics entering the power system and helps to ensure the power quality of the power system.
[0059] Among these measures, harmonic mitigation solutions are obtained for the dominant equipment types. For example, for dominant equipment types with concentrated and high harmonic frequencies, filters of specific frequencies can be installed to directly filter out harmonic currents of the corresponding frequencies, effectively preventing the propagation of harmonics in the power system. For dominant equipment types with more dispersed harmonic distribution, reactive power compensation devices are used to adjust the reactive power distribution of the power system, improve the power factor, and thus indirectly suppress the generation and amplification of harmonics.
[0060] The beneficial effects of this application's embodiments are twofold. Firstly, feature extraction is performed on the harmonic voltage spectrum of electrified railways to obtain a first feature vector, and feature extraction is performed on the harmonic voltage spectrum of new energy inverters to obtain a second feature vector. The first and second feature vectors are concatenated to obtain the harmonic feature vector of the power system. The harmonic feature vector is then processed by a trained deep belief network model to obtain the dominant equipment type of the harmonic source. Since manual acquisition is not required, the acquisition time of the dominant equipment type of the harmonic source is reduced, which is beneficial to improving the acquisition efficiency of the dominant equipment type of the harmonic source. Secondly, since the harmonic source tracing results include the dominant equipment type of the harmonic source, harmonic responsibility can be divided based on the dominant equipment type, which is beneficial to improving the accuracy of harmonic responsibility division.
[0061] Please see Figure 3 , Figure 3 The implementation flowchart of S202 provided in the embodiments of this application is described in detail below: S301, input the fundamental admittance matrix into the power system simulation module. The simulation module uses the Newton-Raphson method to process the fundamental admittance matrix and obtain the effective voltage value of the fundamental component of each node in the power system. S302 uses the harmonic power flow calculation method to process the harmonic admittance matrix, obtains the effective voltage values of the harmonic components of each node in the power system, and generates the total harmonic distortion rate of each node based on the effective voltage values of the harmonic components of each node, the effective voltage values of the fundamental components of each node, and the total harmonic distortion rate model.
[0062] In the embodiments of this application, the Newton-Raphson method adjusts in the direction of the fastest decrease in function value in each iteration, avoiding blind search and reducing the number of invalid iterations.
[0063] For the harmonic source tracing method described in the above embodiments, please refer to [link / reference]. Figure 4 , Figure 4 This is a schematic block diagram of the harmonic source tracing device provided in the embodiments of this application. Figure 4 The harmonic source tracing device 400 shown can be applied to, for example... Figure 1 The application scenario diagram shows electronic devices. The following section uses electronic devices as an example to illustrate this. Figure 4 The harmonic tracing device 400 shown will be described in detail. The harmonic tracing device 400 may include a first acquisition module 401, a generation module 402, a second acquisition module 403, a sampling module 404, and a tracing module 405.
[0064] The first acquisition module 401 is used to acquire the collected data of the power system and construct the fundamental wave admittance matrix and harmonic wave admittance matrix based on the collected data. The generation module 402 is used to process the fundamental admittance matrix using the Newton-Raphson method to obtain the effective voltage value of the fundamental component of each node in the power system, process the harmonic admittance matrix using the harmonic power flow calculation method to obtain the effective voltage value of the harmonic component of each node in the power system, and generate the total harmonic distortion rate of each node based on the effective voltage value of the harmonic component of each node, the effective voltage value of the fundamental component of each node, and the total harmonic distortion rate model. The second acquisition module 403 is used to mark nodes with harmonic distortion rates greater than preset distortion rates as harmonic exceedance points. When the number of harmonic exceedance points is a positive integer, the harmonic emission model of the electrified railway and the harmonic emission model of the new energy inverter are obtained from the simulation file. The sampling module 404 is used to sample multiple parameters of the harmonic source of the power system using the Latin hypercube sampling strategy to obtain multiple sets of parameter combinations. The multiple sets of parameter combinations are processed by the harmonic emission model of the electrified railway to obtain the harmonic data of the electrified railway. The harmonic data of the electrified railway are processed by the three-phase power flow model of the electrified railway to obtain the harmonic voltage spectrum of the electrified railway. The multiple sets of parameter combinations are processed by the harmonic emission model of the new energy inverter to obtain the harmonic data of the new energy inverter. The harmonic data of the new energy inverter are processed by the three-phase power flow model of the new energy inverter to obtain the harmonic voltage spectrum of the new energy inverter. The source tracing module 405 is used to process the harmonic voltage spectrum of the electrified railway using a trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the electrified railway. It also processes the harmonic voltage spectrum of the new energy inverter using the trained deep belief network model to obtain the voltage spectrum amplitude of each harmonic of the new energy inverter. Based on the voltage spectrum amplitudes of each harmonic of the electrified railway, the voltage spectrum amplitudes of each harmonic of the new energy inverter, and the generation model, it generates the root mean square value of the harmonic voltage spectrum of the electrified railway and the harmonic voltage spectrum of the new energy inverter. The root mean square (RMS) value of the harmonic voltage spectrum of the electrified railway is used to identify the dominant equipment type of the harmonic source. When the RMS value of the harmonic voltage spectrum of the electrified railway is greater than the harmonic threshold of the electrified railway and the RMS value of the harmonic voltage spectrum of the new energy inverter is not greater than the harmonic threshold of the new energy inverter, the dominant equipment type of the harmonic source is identified as the new energy inverter.
[0065] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0066] The beneficial effects of this application's embodiments are twofold. Firstly, feature extraction is performed on the harmonic voltage spectrum of electrified railways to obtain a first feature vector, and feature extraction is performed on the harmonic voltage spectrum of new energy inverters to obtain a second feature vector. The first and second feature vectors are concatenated to obtain the harmonic feature vector of the power system. The harmonic feature vector is then processed by a trained deep belief network model to obtain the dominant equipment type of the harmonic source. Since manual acquisition is not required, the acquisition time of the dominant equipment type of the harmonic source is reduced, which is beneficial to improving the acquisition efficiency of the dominant equipment type of the harmonic source. Secondly, since the harmonic source tracing results include the dominant equipment type of the harmonic source, harmonic responsibility can be divided based on the dominant equipment type, which is beneficial to improving the accuracy of harmonic responsibility division.
[0067] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0068] like Figure 5 As shown, Figure 5 The electronic device 2 includes: at least one processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the at least one processor 20, wherein the processor 20 executes the computer program 22 to implement the steps in any of the above method embodiments.
[0069] The electronic device 2 may include, but is not limited to, a processor 20 and a memory 21. Those skilled in the art will understand that... Figure 5 This is merely an example of electronic device 2 and does not constitute a limitation on electronic device 2. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.
[0070] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. A harmonic tracing method based on electrified railway and new energy inverter, characterized in that, The harmonic tracing method is applied to an electronic device, and comprises the following steps: Collecting data of a power system, and constructing a fundamental wave admittance matrix and a harmonic admittance matrix according to the collected data; Processing the fundamental wave admittance matrix by using a Newton-Raphson method to obtain the voltage effective value of the fundamental wave component of each node in the power system, processing the harmonic admittance matrix by using a harmonic power flow calculation method to obtain the voltage effective value of the harmonic component of each node in the power system, and generating the total harmonic distortion rate of each node according to the voltage effective value of the harmonic component of each node, the voltage effective value of the fundamental wave component of each node, and a total harmonic distortion rate model; Marking a node with a harmonic distortion rate greater than a preset distortion rate as a harmonic exceeding point, and when the number of the harmonic exceeding points is a positive integer, obtaining a harmonic emission model of an electrified railway and a harmonic emission model of a new energy inverter from a simulation file; Sampling a plurality of parameters of a harmonic source of the power system by using a Latin hypercube sampling strategy to obtain a plurality of parameter combinations, processing the plurality of parameter combinations by using the harmonic emission model of the electrified railway to obtain harmonic data of the electrified railway, processing the harmonic data of the electrified railway by using a three-phase power flow model of the electrified railway to obtain a harmonic voltage frequency spectrum of the electrified railway, processing the plurality of parameter combinations by using the harmonic emission model of the new energy inverter to obtain harmonic data of the new energy inverter, and processing the harmonic data of the new energy inverter by using a three-phase power flow model of the new energy inverter to obtain a harmonic voltage frequency spectrum of the new energy inverter; Processing the harmonic voltage frequency spectrum of the electrified railway by using a trained deep belief network model to obtain the voltage frequency spectrum amplitude of each harmonic of the electrified railway, processing the harmonic voltage frequency spectrum of the new energy inverter by using the trained deep belief network model to obtain the voltage frequency spectrum amplitude of each harmonic of the new energy inverter, and generating the root mean square value of the harmonic voltage frequency spectrum of the electrified railway and the root mean square value of the harmonic voltage frequency spectrum of the new energy inverter according to the voltage frequency spectrum amplitude of each harmonic of the electrified railway, the voltage frequency spectrum amplitude of each harmonic of the new energy inverter, and a generation model, marking the dominant device type of the harmonic source as the electrified railway when the root mean square value of the harmonic voltage frequency spectrum of the electrified railway is greater than a harmonic threshold value of the electrified railway and the root mean square value of the harmonic voltage frequency spectrum of the new energy inverter is not greater than a harmonic threshold value of the new energy inverter, and marking the dominant device type of the harmonic source as the new energy inverter when the root mean square value of the harmonic voltage frequency spectrum of the electrified railway is not greater than the harmonic threshold value of the electrified railway and the root mean square value of the harmonic voltage frequency spectrum of the new energy inverter is greater than the harmonic threshold value of the new energy inverter.
2. The harmonic traceability method of claim 1, wherein, The method for processing the fundamental wave admittance matrix by using the Newton-Raphson method to obtain the voltage effective value of the fundamental wave component of each node in the power system, processing the harmonic admittance matrix by using the harmonic power flow calculation method to obtain the voltage effective value of the harmonic component of each node in the power system, and generating the total harmonic distortion rate of each node according to the voltage effective value of the harmonic component of each node, the voltage effective value of the fundamental wave component of each node, and the total harmonic distortion rate model comprises the following steps: The fundamental wave admittance matrix is input to a simulation module of the power system, and the simulation module processes the fundamental wave admittance matrix by using a Newton-Raphson method to obtain the effective value of the fundamental wave component of the voltage of each node in the power system; The harmonic wave flow calculation method is used to process the harmonic admittance matrix to obtain the effective value of the harmonic component of the voltage of each node in the power system, and the total harmonic distortion rate model is generated according to the effective value of the harmonic component of the voltage of each node, the effective value of the fundamental wave component of the voltage of each node and the total harmonic distortion rate model.
3. The method of harmonic traceability according to claim 1, characterized in that, The Latin hypercube sampling strategy is used to sample a plurality of parameters of the harmonic source of the power system to obtain a plurality of groups of parameter combinations, the harmonic data of the electrified railway is obtained by processing the plurality of groups of parameter combinations through the harmonic emission model of the electrified railway, the harmonic voltage spectrum of the electrified railway is obtained by processing the harmonic data of the electrified railway through the three-phase power flow model of the electrified railway, the harmonic data of the new energy inverter is obtained by processing the plurality of groups of parameter combinations through the harmonic emission model of the new energy inverter, and the harmonic voltage spectrum of the new energy inverter is obtained by processing the harmonic data of the new energy inverter through the three-phase power flow model of the new energy inverter, including: A plurality of parameters of the harmonic source are obtained from a configuration file, and the Latin hypercube sampling strategy is used to sample the plurality of parameters of the harmonic source of the power system to obtain a plurality of groups of parameter combinations; The plurality of groups of parameter combinations are input to the harmonic emission model of the electrified railway through the first input interface, the harmonic data of the electrified railway is obtained by processing the plurality of groups of parameter combinations through the harmonic emission model of the electrified railway, the harmonic data of the electrified railway is input to the three-phase power flow model of the electrified railway through the second input interface, the harmonic voltage spectrum of the electrified railway is obtained by processing the harmonic data of the electrified railway through the three-phase power flow model of the electrified railway, the plurality of groups of parameter combinations are input to the harmonic emission model of the new energy inverter through the third input interface, the harmonic data of the new energy inverter is obtained by processing the plurality of groups of parameter combinations through the harmonic emission model of the new energy inverter, the harmonic data of the new energy inverter is input to the three-phase power flow model of the new energy inverter through the fourth input interface, and the harmonic voltage spectrum of the new energy inverter is obtained by processing the harmonic data of the new energy inverter through the three-phase power flow model of the new energy inverter.
4. The method of harmonic traceability of claim 1, wherein, The new energy inverter includes a photovoltaic inverter and a wind power inverter.
5. The method of harmonic traceability of claim 1, wherein, The collected data includes topological structure data of the power network, power equipment parameters, parameters of a grid-connected interface of the electrified railway, and parameters of a grid-connected interface of the new energy inverter, and the power equipment parameters include transmission line parameters, transformer parameters and generator parameters.
6. The method of harmonic traceability of claim 1, wherein, The harmonic emission model of the electrified railway is defined as follows: ； wherein, denotes the current amplitude of a traction converter of an electrified railway; Vgrid is the voltage provided by the power grid for the electrified railway; is the power factor, is the power factor angle of the rectifier of the electrified railway; for the traction power of an electrified railway; DC current output by a traction converter of an electrified railway to a traction motor; for the switching frequency of a traction converter of an electrified railway; For industrial frequency; a proportional parameter for a controller of an electrified railway; Integral parameters of the controller of the electrified railway; is a running mode coefficient of the electrified railway; is 1 or 0.8; is 1, indicating that the electrified railway is in a traction state; is 0.8, indicating that the electrified railway is in a braking state; The error signal of the electrified railway is the difference between the current reference value and the actual value in the traction control process of the electrified railway. The error signal of the electrified railway is integrated, aiming at eliminating the steady-state error of the electrified railway, for ensuring stable operation of the traction converter.
7. The harmonic tracing method of claim 1, wherein, The harmonic emission model of the new energy inverter is defined as follows: ； ； represents the harmonic current generated by the new energy inverter; Vdc represents the DC side voltage of the new energy inverter; represents the modulation amplitude value parameter of the new energy inverter; represents the harmonic inductive reactance of the new energy inverter, used to affect the flow of harmonic current; The modulation ratio of the new energy inverter is the ratio of the switching frequency of the new energy inverter to the fundamental frequency of the new energy inverter, and is used to determine the modulation effect of the new energy inverter. The correction term represents a controller of a new energy inverter, and is used for optimizing the control performance of the new energy inverter. represents a switching frequency of the new energy inverter; represents a fundamental frequency of the new energy inverter; represents the fundamental current of the new energy inverter output.
8. The method of harmonic traceability of claim 1, wherein, The generation model is defined as follows: ； The application relates to a method for operating an electrically operated vehicle a voltage spectrum amplitude of the second harmonic; The first harmonic of the voltage spectrum amplitude squared of the electrically The first harmonic of the voltage spectrum amplitude squared of the electrically The first Square of the voltage spectrum amplitude of the second harmonic. The first aspect of the application relates to a new energy inverter the square of the amplitude of the voltage spectrum of the second harmonic RMS value of the harmonic voltage spectrum of the electrified railway; is the root mean square value of the harmonic voltage spectrum of the new energy inverter.
9. The method of harmonic traceability of claim 1, wherein, The total harmonic distortion rate model is defined as follows: ； represents the total harmonic distortion rate of the i-th node, the higher the total harmonic distortion rate of the i-th node, the higher the harmonic distortion degree of the i-th node, the lower the total harmonic distortion rate of the i-th node, the lower the harmonic distortion degree of the i-th node; represents the voltage effective value of the h-th harmonic component of the i-th node, represents the square of the voltage effective value of all harmonic components of the i-th node; represents the voltage effective value of the fundamental component of the i-th node.
10. A harmonic tracing device based on electrified railway and new energy inverter, characterized in that, The application is applied to an electronic device, including: A first obtaining module is configured to obtain collected data of a power system, and construct a fundamental wave admittance matrix and a harmonic admittance matrix according to the collected data. The generating module is configured to process the fundamental wave admittance matrix by using the Newton-Raphson method to obtain the voltage effective value of the fundamental wave component of each node in the power system, process the harmonic admittance matrix by using a harmonic flow calculation method to obtain the voltage effective value of the harmonic component of each node in the power system, and generate the total harmonic distortion rate of each node according to the voltage effective value of the harmonic component of each node, the voltage effective value of the fundamental wave component of each node, and a total harmonic distortion rate model; The second obtaining module is configured to mark a node with a harmonic distortion rate greater than a preset distortion rate as a harmonic over-limit point, and when the number of the harmonic over-limit points is a positive integer, obtain a harmonic emission model of the electrified railway and a harmonic emission model of the new energy inverter from the simulation file; The sampling module is configured to sample a plurality of parameters of the harmonic source of the power system by using a Latin hypercube sampling strategy to obtain a plurality of groups of parameter combinations, process the plurality of groups of parameter combinations by using the harmonic emission model of the electrified railway to obtain harmonic data of the electrified railway, process the harmonic data of the electrified railway by using a three-phase power flow model of the electrified railway to obtain harmonic voltage frequency spectrum of the electrified railway, process the plurality of groups of parameter combinations by using the harmonic emission model of the new energy inverter to obtain harmonic data of the new energy inverter, and process the harmonic data of the new energy inverter by using a three-phase power flow model of the new energy inverter to obtain harmonic voltage frequency spectrum of the new energy inverter; The tracing module is configured to process the harmonic voltage frequency spectrum of the electrified railway by using the trained deep belief network model to obtain voltage frequency spectrum amplitudes of each harmonic of the electrified railway, process the harmonic voltage frequency spectrum of the new energy inverter by using the trained deep belief network model to obtain voltage frequency spectrum amplitudes of each harmonic of the new energy inverter, generate a root mean square value of the harmonic voltage frequency spectrum of the electrified railway and a root mean square value of the harmonic voltage frequency spectrum of the new energy inverter according to the voltage frequency spectrum amplitudes of each harmonic of the electrified railway, the voltage frequency spectrum amplitudes of each harmonic of the new energy inverter, and the generating model, mark the dominant device type of the harmonic source as the electrified railway when the root mean square value of the harmonic voltage frequency spectrum of the electrified railway is greater than a harmonic threshold of the electrified railway and the root mean square value of the harmonic voltage frequency spectrum of the new energy inverter is not greater than a harmonic threshold of the new energy inverter, and mark the dominant device type of the harmonic source as the new energy inverter when the root mean square value of the harmonic voltage frequency spectrum of the electrified railway is not greater than the harmonic threshold of the electrified railway and the root mean square value of the harmonic voltage frequency spectrum of the new energy inverter is greater than the harmonic threshold of the new energy inverter.