A method, system and storage medium for predicting the service life of a linear power supply
By collecting and analyzing the external parameter data of linear power supply, establishing a neural network model, and predicting the ambient temperature of capacitors, solving the accuracy problem of capacitor life prediction in the existing technology, and achieving reliable prediction of the service life of linear power supply.
Patent Information
- Application Number
- CN202110431651.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-04-21
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2041-04-21
AI Technical Summary
The prior art is difficult to accurately predict the service life of capacitors in linear power supplies. Direct measurement of capacitor ambient temperature is invasive and unfeasible, with large errors, which affects the maintenance and maintenance of linear power supplies.
By collecting data on the output voltage, output current, air inlet temperature, air outlet temperature and fan speed of the linear power supply, a neural network model is established to measure and predict the capacitance ambient temperature online, thereby predicting the capacitance life and the service life of the linear power supply.
It realizes accurate estimation of the capacitance ambient temperature without disassembling the power supply, reduces the maintenance of linear power supply, and improves the accuracy and reliability of life prediction.
Smart Images

Figure CN113191410B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of power electronics technology, and in particular to a method, system and storage medium for predicting the service life of a linear power supply. Background Art
[0002] Linear power supplies are widely used in the engineering field due to their advantages such as relatively simple structure, no switching noise, small output ripple, and small high-frequency interference. They play an indispensable role in the field of power electronics, and whether they work normally directly affects the safety of power electronic equipment.
[0003] The capacitor in the linear power supply circuit is the component with the highest failure rate in the entire power supply. There are many factors that cause capacitor failure, among which the linear power supply generates a lot of heat during operation, and the increase in capacitance with temperature is a key factor leading to capacitor failure. The ambient temperature of the capacitor is affected by its own heat, and the heat of surrounding components will also affect the ambient temperature of the capacitor, causing the capacitance of the capacitor to decrease and the equivalent series resistance to increase, which seriously affects the service life of the capacitor. Therefore, the ambient temperature of the capacitor has become one of the most important factors for its failure, and the predicted life of the capacitor is related to the service life of the linear power supply.
[0004] Capacitors are the weak link of linear power supplies and play a vital role in the life of linear power supplies. There are many studies on capacitor fault diagnosis in view of the impact of the ambient temperature of capacitors on the service life of linear power supplies. Through fault diagnosis methods, capacitors can be quickly detected and located after failure, so as to restore the function of linear power supplies as soon as possible. However, there are few studies on methods for predicting the life of capacitors when linear power supplies are running. Life prediction is to predict the time when potential failures occur, which can greatly reduce unnecessary maintenance work. Methods for predicting the life of capacitors include directly measuring the ambient temperature of capacitors to obtain the life. This method can accurately predict the life value, but its disadvantage is that it is difficult to directly measure the ambient temperature of capacitors in an invasive manner during the normal operation of linear power supplies. The methods of measuring the capacitance of capacitors in linear power supplies and measuring the equivalent series resistance of capacitors have low reliability and large errors, making it difficult to predict the life. Summary of the invention
[0005] The present invention provides a method for predicting the service life of a linear power supply, comprising performing the following steps in sequence:
[0006] Step 1: Data collection related quantities; collect the parameters of the linear power supply under various working conditions.
[0007] Step 2: Enter the collected data into the neural network to build a model.
[0008] Step 3: Measure external data online; randomly and synchronously collect the parameters of the linear power supply during the operation of the linear power supply.
[0009] Step 4: Model predicts temperature: The ambient temperature of the capacitor is predicted using the model established in step 2.
[0010] Step 5: Obtain lifespan through temperature; use the ambient temperature of the capacitor predicted in step 4 to predict the lifespan of the capacitor, thereby predicting the service life of the linear power supply.
[0011] As a further improvement of the present invention, in step 1, the following steps are also included:
[0012] Step 10: Connect an adjustable load to the output of the linear power supply.
[0013] Step 11: Determine the acquisition working conditions and adjust different powers to obtain output current and output voltage.
[0014] Step 12: Use a synchronous measuring instrument and a thermocouple to synchronously collect the output voltage, output current, air inlet temperature, air outlet temperature, capacitor ambient temperature of the linear power supply under various working conditions and record the duty cycle of the fan.
[0015] Step 13: Calculate the fan speed using the fan duty cycle.
[0016] As a further improvement of the present invention, after step 13, the following steps are further performed:
[0017] Step 14: Preprocess the data collected under various working conditions, mainly to eliminate wild points in the measurement data, which is conducive to the establishment of the neural network model.
[0018] As a further improvement of the present invention, in step 2, the measured data is used as a training sample of the neural network, the input is the output voltage, current, air inlet temperature, air outlet temperature and fan speed of the linear power supply under various working conditions, and the output is the ambient temperature of the capacitor to train the neural network.
[0019] As a further improvement of the present invention, in step 3, the following steps are specifically further included:
[0020] Step 30: During the operation of the linear power supply, the output voltage, output current, air inlet temperature, air outlet temperature of the linear power supply are randomly and synchronously collected, and the duty cycle of the fan is recorded.
[0021] Step 31: Calculate the fan speed using the fan duty cycle.
[0022] As a further improvement of the present invention, after step 31, the following steps are further performed:
[0023] Step 32: Preprocess the collected data, mainly removing wild points in the measurement data to ensure the reliability of the measurement data.
[0024] As a further improvement of the present invention, in step 4, the following steps are also included:
[0025] Step 40: Read the measured data into the trained neural network as input.
[0026] Step 41: The output obtained through the neural network is the ambient temperature of the capacitor.
[0027] The present invention also discloses a system for predicting the service life of a linear power supply, comprising: a memory, a processor, and a computer program stored in the memory, wherein the computer program is configured to implement the steps of the method of the present invention when called by the processor.
[0028] The present invention also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps of the method described in the present invention when called by a processor.
[0029] The beneficial effects of the present invention are as follows: 1. The method of the present invention is applicable to various linear power supplies, has a wide range of applications, and is highly applicable; 2. The method of the present invention can establish a nonlinear relationship between the output voltage, output current, air inlet temperature, air outlet temperature, and fan speed and the capacitor during the operation of the linear power supply, and can well estimate the ambient temperature of the capacitor; 3. The method of the present invention does not require the power supply to be disassembled during use, and only requires measuring the external data of the linear power supply to obtain the ambient temperature of the capacitor inside the power supply; 4. The method of the present invention only requires obtaining data of relevant parameters, does not require analyzing the internal structure of the linear power supply, and will not affect the use performance and working state of the linear power supply; 5. The method of the present invention estimates the ambient temperature of the capacitor and is applicable to capacitor life estimation, thereby achieving the purpose of linear power supply service life prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 is a graph of predicted temperature values and actual temperature values of the method of the present invention;
[0031] Figure 2 is an error diagram of the predicted temperature value and the actual temperature value of the method of the present invention;
[0032] Figure 3 is a flow chart of the method of the present invention. DETAILED DESCRIPTION
[0033] like Figure 3As shown, the present invention discloses a method for predicting the service life of a linear power supply. The method only needs to establish a model based on measurement data, measure the output current and voltage values, the temperatures of the air inlet and the air outlet, and the speed of the fan online, estimate the ambient temperature of the capacitor through these parameters and the model, and use the ambient temperature to predict the life of the capacitor, thereby achieving the purpose of predicting the service life of the linear power supply.
[0034] The present invention discloses a method for predicting the service life of a linear power supply, comprising sequentially performing the following steps:
[0035] Step 1: Data collection related quantities; collect the parameters of the linear power supply under various working conditions.
[0036] In step 1, the following steps are also included:
[0037] Step 10: Connect an adjustable load to the output of the linear power supply;
[0038] Step 11: Determine the acquisition working conditions and adjust different powers to obtain output current and output voltage;
[0039] Step 12: Use synchronous measuring instruments, thermocouples and other instruments to synchronously collect the output voltage, current, air inlet temperature, air outlet temperature, capacitor ambient temperature of the linear power supply under various working conditions and record the duty cycle of the fan;
[0040] Step 13: Calculate the fan speed using the fan duty cycle;
[0041] Step 14: Preprocess the collected data under various working conditions, mainly to remove wild points in the measured data, which is conducive to the establishment of the neural network model;
[0042] Step 2: Enter the collected data into the neural network to build a model.
[0043] In step 2, the measured data is used as a training sample of the neural network, the input is the output voltage, output current, air inlet temperature, air outlet temperature and fan speed of the linear power supply under various working conditions, and the output is the ambient temperature of the capacitor to train the neural network.
[0044] Step 3: Measure external data online; randomly and synchronously collect the parameters of the linear power supply during the operation of the linear power supply.
[0045] In step 3, the following steps are also specifically performed:
[0046] Step 30: During the operation of the linear power supply, the output voltage, output current, air inlet temperature, air outlet temperature of the linear power supply are randomly and synchronously collected, and the duty cycle of the fan is recorded;
[0047] Step 31: Calculate the fan speed using the fan duty cycle;
[0048] Step 32: Preprocess the collected data, mainly removing wild points in the measurement data to ensure the reliability of the measurement data.
[0049] Step 4: Model predicts temperature: The ambient temperature of the capacitor is predicted using the model established in step 2.
[0050] In step 4, the following steps are also included:
[0051] Step 40: Read the measured data into the trained neural network as input;
[0052] Step 41: The output obtained through the neural network is the ambient temperature of the capacitor.
[0053] Step 5: Obtain lifespan through temperature; use the ambient temperature of the capacitor predicted in step 4 to predict the lifespan of the capacitor, thereby predicting the service life of the linear power supply.
[0054] The working principle of the method for predicting the service life of a linear power supply of the present invention is as follows:
[0055] The ambient temperature of the capacitor in the linear power supply is related to the output voltage, output current, air inlet temperature, air outlet temperature and fan speed. The measured data of the variables is used as the input of the prediction model to train the neural network, and the ambient temperature of the capacitor in the linear power supply is used as the output. The life of the capacitor is predicted based on the relationship between the ambient temperature of the capacitor and the life of the capacitor obtained by the prediction model, thereby achieving the purpose of linear power supply life prediction.
[0056] The sample data provided by the present invention comes from ten sets of experimental measurement results of linear power supplies. The specific parameter measurements are shown in Table 1. Seven sets of data are randomly used as historical data to train the model, and the output data obtained by inputting variables of three sets of data are used as predicted values. The predicted temperature is compared with the actual temperature. Figure 1 As shown, it is used to verify the reliability of the trained neural network. The error between the predicted temperature and the actual temperature is as follows Figure 2 As shown, the method of the present invention can accurately predict the ambient temperature of the capacitor.
[0057] Table 1 Related measurement data
[0058]
[0059] The present invention also discloses a system for predicting the service life of a linear power supply, comprising: a memory, a processor, and a computer program stored in the memory, wherein the computer program is configured to implement the steps of the method of the present invention when called by the processor.
[0060] The present invention also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps of the method described in the present invention when called by a processor.
[0061] The beneficial effects of the present invention are as follows: 1. The method of the present invention is applicable to various linear power supplies, has a wide range of applications, and is highly applicable; 2. The method of the present invention can establish a nonlinear relationship between the output voltage, output current, air inlet temperature, air outlet temperature, and fan speed and the ambient temperature of the capacitor during the operation of the linear power supply, and can well estimate the ambient temperature of the capacitor; 3. The method of the present invention does not require the power supply to be disassembled during use, and only requires measuring the external data of the linear power supply to obtain the ambient temperature of the capacitor inside the power supply; 4. The method of the present invention only requires obtaining data of relevant parameters, and does not require analyzing the internal structure of the linear power supply, and will not affect the use performance and working state of the linear power supply; 5. The method of the present invention estimates the ambient temperature of the capacitor and is suitable for capacitor life estimation, thereby achieving the purpose of linear power supply service life prediction.
[0062] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.
Claims
1. A method for predicting the service life of a linear power supply, characterized in that: It includes performing the following steps in sequence: Step 1: Data collection related quantities: Collect the parameters of the linear power supply under various working conditions; specifically, perform the following steps: Step 10: Connect an adjustable load to the output of the linear power supply; Step 11: Determine the acquisition working conditions and adjust different powers to obtain output current and output voltage; Step 12: Use a synchronous measuring instrument and a thermocouple to synchronously collect the output voltage, current, air inlet temperature, air outlet temperature, capacitor ambient temperature of the linear power supply under various working conditions and record the duty cycle of the fan; Step 13: Calculate the fan speed using the fan duty cycle; Step 14: Preprocess the data collected under various working conditions and remove wild points in the measured data; Step 2: Input the collected data into the neural network to establish a model, specifically including: using the measured data as a training sample of the neural network, the input is the output voltage, output current, air inlet temperature, air outlet temperature and fan speed of the linear power supply under various working conditions, and the output is the ambient temperature of the capacitor, and the neural network model is trained; Step 3: Online measurement of external data: Randomly and synchronously collect the parameters of the linear power supply during the operation of the linear power supply, specifically including the following steps: Step 30: During the operation of the linear power supply, the output voltage, output current, air inlet temperature, air outlet temperature of the linear power supply are randomly and synchronously collected, and the duty cycle of the fan is recorded; Step 31: Calculate the fan speed using the fan duty cycle; Step 4: Model prediction temperature: predict the ambient temperature of the capacitor using the model established in step 2; Step 5: Obtaining life by temperature: using the ambient temperature of the capacitor predicted in step 4 to predict the life of the capacitor, thereby predicting the service life of the linear power supply; Wherein, the step 4 specifically includes executing the following steps: Step 40: Input the measured data into the trained neural network model as input; Step 41: The output obtained by the neural network model is the ambient temperature of the capacitor.
2. The method according to claim 1, characterized in that After step 31, the method further includes performing the following steps: Step 32: Preprocess the collected data and remove wild points in the measurement data to ensure the reliability of the measurement data.
3. A system for predicting the service life of a linear power supply, characterized in that: include: A memory, a processor and a computer program stored in the memory, wherein the computer program is configured to implement the steps of the method according to claim 1 or 2 when called by the processor.
4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps of the method according to claim 1 or 2 when called by a processor.
