A method for diagnosing the state of a reactor based on vibration analysis

Abstract

The application provides a kind of reactive device state diagnosis method based on vibration analysis, establishes the finite element analysis model of reactive device for oil-immersed reactor;Establish a correlation influence analysis strategy, including the action relationship and influence relationship between the correlation vibration influence information and the finite element analysis model;Get the real-time vibration information of the actual reactive device and use it as the vibration influence input information of the finite element analysis model;According to the correlation influence analysis strategy, analyze and present the vibration influence input information through the finite element analysis model;Compare the analysis result with the preset threshold value, and make state diagnosis and remind according to the comparison result.The application establishes the finite element analysis model of oil-immersed reactor according to the structure and characteristics of oil-immersed reactor, obtains the vibration characteristics of oil-immersed reactor, realizes the online monitoring and state evaluation, trend analysis and abnormal early warning of the vibration information of reactive device winding and core, and guarantees the safe and reliable operation of oil-immersed reactor and power system.

Description

Technical Field

[0001] This invention relates to the field of power distribution networks, and in particular to a method for reactor condition diagnosis based on vibration analysis. Background Technology

[0002] Reactors, also known as inductors, are widely used in high-voltage power transmission systems for reactive power compensation and possess excellent operating characteristics. However, according to fault statistics, the failure rate of reactors is significantly higher than that of main transformers of the same voltage level. These failures primarily stem from winding deformation, inter-turn short circuits, and loosening of the core and its clamping components. Furthermore, core component failures are also a major cause of failure in cored reactors. When the magnetic circuit of the reactor core malfunctions, or when the core is not properly clamped during manufacturing and installation, or when clamping components are loose, core vibration is significantly aggravated, further leading to loosening of fasteners. In severe cases, this can result in multi-point grounding faults in the core and partial discharge caused by accessories contacting the casing.

[0003] Currently, monitoring of reactors generally relies on sensors to monitor the overall vibration of the reactor. This method cannot reflect the condition of the windings and core within the reactor, and only alerts staff when the vibration becomes severe or a fault has already occurred. Due to the lack of effective monitoring technology, such faults occur frequently, posing a serious threat to the operational safety of the power system. Summary of the Invention

[0004] To address the problems existing in the background art, this invention proposes a reactor condition diagnosis method based on vibration analysis.

[0005] A reactor condition diagnosis method based on vibration analysis is proposed, which involves establishing a finite element analysis model of the oil-immersed reactor; establishing a correlation influence analysis strategy, including the interaction and influence relationships between vibration influence information and the finite element analysis model; acquiring real-time vibration information of the actual reactor and using it as vibration influence input information for the finite element analysis model; analyzing and presenting the vibration influence input information through the finite element analysis model according to the correlation influence analysis strategy; comparing the analysis results with preset thresholds, and performing condition diagnosis and alerts based on the comparison results.

[0006] Based on the above, the finite element analysis model of the reactor includes the winding model and the core model.

[0007] Based on the above, the frequency of change of the reactor input current is obtained, and twice the frequency of change of the input current is used as a vibration influence input information for the iron core model.

[0008] Based on the above, the magnetic field distribution of the reactor is obtained and the corresponding magnetization intensity distribution is calculated. The magnetostrictive deformation effect of the core model is calculated according to the properties established by the core model corresponding to the actual core, and is used as another vibration influence input information for the core model.

[0009] Based on the above, the vibration information of the reactor is obtained, and the vibration information after filtering out the vibration influence input information of the core model is used as the vibration influence input information of the winding model.

[0010] Based on the above, the vibration impact input information of the core model and winding model is detected respectively, and an early warning is issued when the real-time vibration impact input information exceeds the preset threshold.

[0011] Based on the above, the vibration impact input information is recorded to show the real-time and cumulative impact information on the core model and winding model, and early warning is given based on the cumulative deformation of the core model and winding model.

[0012] Based on the above, a historical database is established to record and store the vibration information of the reactor over time. Based on the historical periodic data and the cumulative deformation of the core model and winding model, vibration trends and fault warnings are predicted.

[0013] This invention has outstanding substantive features and significant progress compared to the prior art. Specifically, this invention establishes a finite element analysis model of oil-immersed reactors based on their structure and characteristics, obtains the vibration characteristics of oil-immersed reactors and their influencing factors, and realizes online monitoring, status assessment, trend analysis and anomaly early warning of vibration information of the windings and cores of the oil-immersed reactors in operation, so as to ensure the safe and reliable operation of power oil-immersed reactors and power systems. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the process of this invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] like Figure 1As shown, a reactor condition diagnosis method based on vibration analysis is proposed. This method establishes a finite element analysis model of the oil-immersed reactor; establishes a correlation influence analysis strategy, including the interaction and influence relationships between vibration influence information and the finite element analysis model; acquires real-time vibration information of the actual reactor and uses it as vibration influence input information for the finite element analysis model; analyzes and presents the vibration influence input information through the finite element analysis model according to the correlation influence analysis strategy; compares the analysis results with preset thresholds, and performs condition diagnosis and alerts based on the comparison results.

[0017] Specifically, a finite element analysis (FEM) model of the actual oil-immersed reactor is established, corresponding to its components, attributes, and interrelationships. Based on mechanical dynamics analysis, a correlation influence analysis strategy is developed. The action relationship refers to establishing how vibration influence information acts on the finite element analysis model, while the influence relationship refers to establishing the mechanism by which vibration influence information affects the attributes and changes of the finite element analysis model. After acquiring real-time vibration information of the actual reactor, it is input into the finite element analysis model as vibration influence input information. Following the correlation influence analysis strategy, the vibration influence input information is correlated with the finite element analysis model. The state of the finite element analysis model is then analyzed and observed. Based on the comparison of the analysis results with thresholds, the vibration influence of the finite element analysis model is diagnosed and alerted, thus reflecting the vibration influence state of the actual oil-immersed reactor and achieving real-time online monitoring and state assessment of the actual oil-immersed reactor.

[0018] In this embodiment, the monitoring objects of the finite element analysis model of the reactor mainly include the winding model and the core model. With the use of high-permeability silicon steel sheets in reactor manufacturing and the improvement of core structure design, the working magnetic flux density of the core has decreased, and the winding vibration caused by leakage flux generated by the load current has also increased significantly. The winding vibration is close to, or even exceeds, the core vibration caused by the magnetostriction of silicon steel sheets. Therefore, the vibration of modern reactors mainly depends on the vibration of the windings and the core.

[0019] The vibration of the reactor core during stable operation is mainly caused by two factors: First, the core vibration is caused by the Maxwell force between the core plates. The reactor core is segmented, with many air gaps, leading to an unbalanced magnetic flux distribution across the core cross-section. Under the influence of an alternating electromagnetic field, Maxwell forces are generated between the core plates. These forces cause the magnetic field energy to shift towards the weaker direction, resulting in elastic deformation of the core. Due to the characteristics of the Maxwell force, which changes with the current, the attraction between the core plates of the UHV parallel reactor has the following characteristic: the frequency of change is twice the power supply frequency. In this embodiment, the frequency of the reactor input current is obtained, and twice the frequency of the input current change is used as the vibration influence input information of the core model.

[0020] Another reason is the core vibration caused by the magnetostriction of the silicon steel sheets in the core. Magnetostriction causes periodic vibrations in the core of the UHV parallel reactor. When the reactor is working, a magnetic circuit flows through the core. Due to the air gap between the core discs, electromagnetic forces are generated on the upper and lower sides of the core discs. Along the direction of the magnetic circuit, the core undergoes magnetostrictive deformation. Core vibration is the root cause of vibration and noise in UHV parallel reactors. In this embodiment, the magnetic field distribution of the reactor is obtained and the corresponding magnetization intensity distribution is calculated. Based on the properties established by the core model corresponding to the actual core, the magnetostrictive deformation effect of the core model is calculated and used as another vibration influence input information for the core model.

[0021] Reactor winding deformation refers to irreversible changes in the size or shape of the winding under mechanical or electrodynamic forces. Examples include changes in axial or radial dimensions, reactor body displacement, winding twisting, bulging, and inter-turn short circuits. If a reactor winding is severely deformed and continues to operate without diagnosis, it is highly likely to cause an accident, ranging from power outages to reactor burnout. Besides manufacturing defects or unreasonable factors, the main causes of reactor winding deformation are the influence of internal electrodynamic forces and external mechanical forces, with electrodynamic forces having the most significant impact. For example, the short-circuit impact current and resulting electrodynamic force from a short circuit at the reactor outlet can cause winding twisting, deformation, or even collapse. Since winding vibration is influenced by many factors and is difficult to diagnose, this embodiment acquires the reactor's vibration information and uses the vibration information after filtering out the vibration influence input information of the core model as the vibration influence input information of the winding model.

[0022] The vibration impact input information of the core model and winding model is detected and acquired separately, and the change threshold is preset for each. When the real-time vibration impact input information exceeds the preset threshold, an early warning is issued to avoid damage caused by excessive vibration.

[0023] Preferably, the vibration impact input information is recorded in both real-time and cumulatively on the core and winding models. That is, based on the impact of vibration on the properties and changes of the winding and core, and the resulting state changes after a certain period of accumulation, the model with the accumulated changes is used as the monitoring object, rather than the original model. Under real-time vibration impact information, early warnings are issued based on the cumulative deformation of the core and winding models.

[0024] In practice, a historical database is also established to record and store the vibration information of the reactor over time. Based on the historical periodic data and the cumulative deformation of the core model and winding model, vibration trends and fault warnings are predicted.

[0025] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A reactor condition diagnosis method based on vibration analysis, characterized in that: Establish a finite element analysis model for the corresponding oil-immersed reactor; Establish a correlation impact analysis strategy, including the interaction and influence relationships between vibration impact information and finite element analysis models; The real-time vibration information of the actual reactor is obtained and used as the vibration influence input information for the finite element analysis model; Based on the correlation impact analysis strategy, the vibration impact input information is analyzed and presented using a finite element analysis model; The analysis results are compared with preset thresholds, and status diagnosis and reminders are given based on the comparison results; The finite element analysis model of a reactor includes a winding model and a core model; The frequency of the input current of the reactor is obtained, and twice the frequency of the input current change is used as a vibration influence input information of the iron core model. Obtain the magnetic field distribution of the reactor and calculate the corresponding magnetization intensity distribution. Calculate the magnetostrictive deformation effect of the core model based on the properties established by the core model corresponding to the actual core, and use this as another vibration influence input information for the core model. The vibration information of the reactor is obtained, and the vibration information after filtering out the vibration influence input information of the core model is used as the vibration influence input information of the winding model.

2. The reactor condition diagnosis method based on vibration analysis according to claim 1, characterized in that: The vibration impact input information of the core model and winding model is detected separately, and an early warning is issued when the real-time vibration impact input information exceeds the preset threshold.

3. The reactor condition diagnosis method based on vibration analysis according to claim 1, characterized in that: Record the real-time and cumulative impact information of vibration on the core model and winding model, and provide early warnings based on the cumulative deformation of the core model and winding model.

4. The reactor condition diagnosis method based on vibration analysis according to claim 3, characterized in that: Establish a historical database to record and store the vibration information of the reactor over time. Based on the historical periodic data and the cumulative deformation of the core model and winding model, predict vibration trends and provide early warning of faults.

Images