A medium voltage switch cabinet operation state evaluation method based on multi-parameter time domain analysis

Abstract

The application discloses a medium-voltage switch cabinet operation state evaluation method based on multi-parameter time domain analysis. The method monitors switch cabinet operation temperature and multiple switch cabinet partial discharge data, and constructs a time domain analysis model for the temperature and the multiple partial discharge data. The method comprehensively evaluates the medium-voltage switch cabinet state by using real-time monitoring data and predicted data of the time domain analysis model. The application establishes a predicted trend of key operation monitoring parameters of the switch cabinet, early discovers state changes of the switch cabinet, and warns operation and maintenance personnel, so that hidden dangers in dangerous operation can be early eliminated.

Description

Technical Field

[0001] This invention relates to the field of power technology, specifically a method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis. Background Technology

[0002] With the development of my country's power industry, the scale of the power grid is increasing daily. Medium-voltage switchgear, as a crucial unit of the power system, is a vital link in ensuring the normal operation of the power grid. Its main function is to switch, control, and protect electrical equipment during power generation, transmission, distribution, and energy conversion. It has become one of the most numerous and widely used power switching devices. Medium-voltage switchgear in the power grid is characterized by its large quantity and high failure rate. Its complex structure and confined internal space make it susceptible to damage from environmental factors and various physical and chemical effects during long-term operation. The probability of failure due to mechanical damage, insulation moisture, and material defects is increasing, potentially damaging equipment or even causing serious power outages and significant economic losses.

[0003] Currently, the maintenance of medium-voltage switchgear equipment mainly relies on periodic inspections, but this method has certain drawbacks. On the one hand, there is a shortage of maintenance personnel, and overdue maintenance is common. On the other hand, some operational status monitoring methods are limited, and errors during maintenance work may endanger the personal safety of maintenance personnel. Therefore, online monitoring of switchgear operational status is of great significance. However, existing medium-voltage switchgear status monitoring solutions have low sensitivity, are prone to false alarms, and cannot predict switchgear degradation in advance, exhibiting a certain degree of lag. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis. This method integrates sensing data from multiple sensors and establishes a prediction of the changing trends of partial discharge and temperature data based on a time-domain analysis model. It analyzes the current data and the model prediction data to achieve early detection of dangerous factors and early warning in the early stage of switchgear deterioration.

[0005] To solve the aforementioned technical problem, the technical solution adopted by this invention is: a method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis, comprising the following steps: S01) Obtain historical data of the medium-voltage switchgear over a period of time and construct a p-order time-domain analysis model: , in Let be the predicted value at time t. , , , before time t Historical data, order Indicates the predicted length. The constant coefficient, For random disturbance terms; S02) Place the time-domain analysis model on the switchgear status analysis platform. Switchgear temperature, ultrasonic partial discharge, ground wave partial discharge, and ultra-high frequency partial discharge each correspond to a time-domain analysis model. S03) The switchgear status analysis platform receives temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device, inputs the monitoring data into the corresponding time domain analysis model, and obtains the switchgear status prediction data for the next moment. S04) Evaluate the status of medium-voltage switchgear based on the temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device, and the prediction data derived from the time-domain analysis model. Specifically: If any of the three partial discharge monitoring data—ultrasound, ground wave, and ultra-high frequency—exceeds the warning threshold, the sampling frequency of the partial discharge acquisition device is increased; if two or more partial discharge monitoring data exceed the warning threshold, the insulation status of the medium-voltage switchgear is considered to have a problem. If none of the three partial discharge monitoring data exceed the warning threshold, the predicted data obtained from the time domain analysis model is analyzed. If two or more partial discharge predicted data exceed the warning threshold, the partial discharge of the medium-voltage switchgear is determined to be in an abnormal state. If the current temperature monitoring data or the temperature prediction value obtained through the time-domain prediction model exceeds the warning threshold, the temperature is considered to be abnormal.

[0006] Furthermore, constant coefficients are estimated using the least squares method. , It is by The specific process of forming the matrix is ​​as follows: For the sample ,when When the white noise is, the estimate is: , in The parameters are The estimation of least squares method aims to minimize the sum of squared residuals, i.e. make , achievable , Substitute the sum of squared residuals and adjust the parameters Taking the derivative and setting it to 0, we get: , therefore The least squares estimate of the parameters is: .

[0007] Furthermore, It is a white noise disturbance.

[0008] Furthermore, after the time-domain analysis model has been running for a period of time, the constant coefficients are optimized by collecting the running data.

[0009] Furthermore, the temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device are preprocessed before being input into the time-domain analysis model. This preprocessing includes outlier analysis and removal, and calculation of the effective value of the partial discharge signal. Outlier analysis and removal takes into account the problem of acquisition interference; obviously abnormal data is removed based on experience to ensure the accuracy of the model. Calculation of the effective value of the partial discharge signal considers the characteristics of high-frequency partial discharge signals. Each time a partial discharge signal is acquired, the average amplitude of the partial discharge signal over a period of time is calculated as the input to the time prediction model. It should be noted that there are many characteristic quantities of the partial discharge signal, including the number of discharges, the peak discharge value, and the average discharge value, all of which can be used as features for evaluating the partial discharge signal.

[0010] The beneficial effects of this invention are as follows: This invention proposes a switchgear condition assessment method based on a multi-parameter time-domain prediction model, which aims to assess the condition of medium-voltage switchgear. By establishing the prediction trend of key operation monitoring parameters of the switchgear, the method can detect changes in the switchgear condition in advance and issue warnings to operation and maintenance personnel, thereby eliminating potential hazards in the early stages of dangerous operation. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the principle of the present invention. Detailed Implementation

[0012] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0013] Example 1 This embodiment addresses the status monitoring problem of medium-voltage switchgear. It monitors temperature and partial discharge data through partial discharge and temperature monitoring devices and uploads the monitoring data. Using the method proposed in this invention, the status of the switchgear is predicted and evaluated, helping operators to detect faults early and eliminate potential hazards.

[0014] The switchgear monitoring device involved in this embodiment is a partial discharge and temperature monitoring device. The partial discharge monitoring device includes monitoring methods such as ultrasonic waves, ground waves, and ultra-high frequency (UHF). The device is magnetically attached to the switchgear cabinet and uploads data to the analysis platform periodically via wireless transmission methods such as LoRa. The temperature monitoring device is installed inside the switchgear and also uploads temperature data via LoRa or similar methods. The partial discharge and temperature acquisition devices periodically collect relevant data from the switchgear and upload it.

[0015] The switchgear status analysis platform has built-in time-domain analysis models for different parameters. The time-domain analysis model is modeled using an autoregressive model (AR). By establishing a time-domain predictive monitoring model for the switchgear under multiple parameters, it can predict the operational change trend of the switchgear in a timely manner and evaluate the operational status of the switchgear by combining existing monitoring data.

[0016] like Figure 1 As shown, the steps for model building and switchgear status analysis are as follows: Step 1: Collect partial discharge and temperature operating data of the switchgear over a period of time, remove unreasonable data, preprocess the data, and construct a p-order time-domain analysis model: , in Let be the predicted value at time t. , , , Before time t Historical data, order Indicates the predicted length. The constant coefficient, This is a random disturbance term.

[0017] Based on this time-domain analysis model, it can be seen that... The value at time passed through the previous The number of time periods is predicted. In this invention, the random disturbance term is considered as a white noise disturbance.

[0018] This embodiment estimates the constant coefficients using the least squares method. , It is by The specific process of forming the matrix is ​​as follows: For the sample ,when When the white noise is, the estimate is: , in The parameters are The estimation of least squares method aims to minimize the sum of squared residuals, i.e. make , We can obtain: , Substitute the sum of squared residuals and adjust the parameters Taking the derivative and setting it to 0, we get: , therefore The least squares estimate of the parameters is: .

[0019] Step 2: Place the time-domain analysis model on the switchgear status analysis platform. Switchgear temperature, ultrasonic partial discharge, ground wave partial discharge, and ultra-high frequency partial discharge each correspond to a time-domain analysis model.

[0020] Step 3: The switchgear status analysis platform receives real-time temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device, removes abnormal data, and inputs the normal data into the time domain analysis model to obtain the switchgear status prediction data for the next moment.

[0021] Step 4: Evaluate the status of the medium-voltage switchgear based on the temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device, and the prediction data derived from the time-domain analysis model. Specifically: If any of the three partial discharge monitoring data—ultrasound, ground wave, and ultra-high frequency—exceeds the warning threshold, the sampling frequency of the partial discharge acquisition device is increased; if two or more partial discharge monitoring data exceed the warning threshold, the insulation status of the medium-voltage switchgear is considered to have a problem. If none of the three partial discharge monitoring data exceed the warning threshold, the predicted data obtained from the time domain analysis model is analyzed. If two or more partial discharge predicted data exceed the warning threshold, the partial discharge of the medium-voltage switchgear is determined to be in an abnormal state. If the current temperature monitoring data or the temperature prediction value obtained through the time-domain prediction model exceeds the warning threshold, the temperature is considered to be abnormal.

[0022] Regarding switchgear monitoring, if the aforementioned abnormal phenomena occur in partial discharge or temperature, the switchgear is considered to be in a dangerous operating state. The system will issue a warning to the maintenance personnel to promptly investigate and eliminate the dangerous factors.

[0023] Step 5: Considering that different switchgear have different operating states, after the model has been running for a period of time, the operating data is collected, and the parameters of the time-domain prediction model are optimized using Step 1 to make it more consistent with the current operating conditions.

[0024] The above description is merely the basic principle and preferred embodiment of the present invention. Improvements and substitutions made by those skilled in the art based on the present invention are within the scope of protection of the present invention.

Claims

1. A method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis, characterized in that: Includes the following steps: S01) Obtain historical data of the medium-voltage switchgear over a period of time and construct a p-order time-domain analysis model: ， in Let be the predicted value at time t. , , , Before time t Historical data, order Indicates the predicted length. The constant coefficient, For random disturbance terms; S02) Place the time-domain analysis model on the switchgear status analysis platform. Switchgear temperature, ultrasonic partial discharge, ground wave partial discharge, and ultra-high frequency partial discharge each correspond to a time-domain analysis model. S03) The switchgear status analysis platform receives temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device, inputs the monitoring data into the corresponding time domain analysis model, and obtains the switchgear status prediction data for the next moment. S04) Evaluate the status of medium-voltage switchgear based on the temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device, and the prediction data derived from the time-domain analysis model. Specifically: If any of the three partial discharge monitoring data—ultrasound, ground wave, and ultra-high frequency—exceeds the warning threshold, the sampling frequency of the partial discharge acquisition device is increased; if two or more partial discharge monitoring data exceed the warning threshold, the insulation status of the medium-voltage switchgear is considered to have a problem. If none of the three partial discharge monitoring data exceed the warning threshold, the predicted data obtained from the time domain analysis model is analyzed. If two or more partial discharge predicted data exceed the warning threshold, the partial discharge of the medium-voltage switchgear is determined to be in an abnormal state. If the current temperature monitoring data or the temperature prediction value obtained through the time-domain prediction model exceeds the warning threshold, the temperature is considered to be abnormal.

2. The method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis according to claim 1, characterized in that: Estimating constant coefficients using the least squares method , It is by The specific process of forming the matrix is ​​as follows: For the sample ,when When the white noise is, the estimate is: ， in The parameters are The estimation of least squares method aims to minimize the sum of squared residuals, i.e. make ， achievable ， Substitute the sum of squared residuals and adjust the parameters Taking the derivative and setting it to 0, we get: ， therefore The least squares estimate of the parameters is: 。 3. The method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis according to claim 1, characterized in that: It is white noise perturbation.

4. The method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis according to claim 2, characterized in that: After the time-domain analysis model has been running for a period of time, the constant coefficients are optimized by collecting the running data.

5. The method for evaluating the operating status of medium-voltage switchgear based on multi-parameter time-domain analysis according to claim 1, characterized in that: The temperature and partial discharge monitoring data uploaded by the temperature acquisition device and the partial discharge acquisition device are first preprocessed before being input into the time domain analysis model. The preprocessing includes outlier analysis and removal, and calculation of the effective value of the partial discharge signal.

Images