Laser power supply assembly performance detection method and system
Through the combination of LOF algorithm and BP neural network model, abnormal data are screened and performance indicators are corrected, which solves the deviation problem of external influencing factors in the existing technology on the dynamic power test of laser power, and improves the accuracy of measurement results.
Patent Information
- Application Number
- CN202411561878.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-05
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-11-05
Smart Images

Figure CN119064820A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of laser power supply detection, and in particular to a method and system for detecting performance of a laser power supply component. Background Art
[0002] Laser power supply is a device that can convert electrical energy into light energy. It is widely used in laser communication, laser display, laser cutting, laser welding and other fields. In order to ensure the performance and reliability of laser power supply, a series of performance tests including static power test, dynamic power test, efficiency test, stability test, noise test, etc. are required.
[0003] Among them, dynamic power test is a method to evaluate the dynamic power performance of laser power supply. At present, dimmer is usually used for measurement. During the test, the laser power supply is connected to the dimmer and power meter, the brightness of the dimmer is dynamically adjusted, and the brightness reading and power meter reading are recorded. The rated power, actual power, efficiency and other performance indicators of the power supply are evaluated based on the obtained brightness and power values.
[0004] However, the existing test methods do not take into account the influence of external factors, such as environmental interference and battery status, on the measurement process, resulting in a deviation between the measurement results and the actual situation, and cannot truly reflect the dynamic power performance of the laser power supply being tested. To this end, we propose a laser power supply component performance detection method and system. Summary of the invention
[0005] The main purpose of the present invention is to provide a method and system for detecting the performance of a laser power supply assembly, which can effectively solve the problems in the background technology.
[0006] In order to achieve the above object, the technical solution adopted by the present invention is: A method for detecting performance of a laser power supply assembly, comprising: Connect the power supply to be tested to the dimmer and power meter, dynamically adjust the brightness of the dimmer, and set the power supply to be tested from the upper limit of the state of charge SOC max Continue to discharge to the lower limit of charge state SOC min The performance index parameters and the brightness value, environmental parameters and state of charge value of the dimmer during the process are sampled, wherein the performance index parameters include rated power, actual power and conversion efficiency; the environmental parameters include the operating temperature of the power supply to be tested and the humidity of the environment; the upper limit SOC of the state of charge max The value range is [0.8,0.9], and the lower limit of the state of charge SOC min The value range of is [0.15,0.2]; The LOF algorithm is used to calculate the local outlier factor in the performance indicator parameter, and the jth sampling value of the i-th performance indicator is calculated using the obtained local outlier factor. Whether it is abnormal data is determined, and abnormal data in the performance indicator parameter sampling values is screened and removed. The determination process includes the following steps: Calculate the sample values separately With the rest of the sample values The distance between , the calculation formula is: Where j, q∈N, and q≠j, N is the total amount of sampled data; the distance value obtained Sort by small to large, set and sample value The kth element closest to p∈N, and p≠j, get the sample value With the sample value The kth distance value ,in, ; According to the kth distance value Identify the elements The kth distance neighbor of ; Get sample value The k-th reachable distance of , is the sampling value The k-th reachable distance of To obtain , The maximum value operation in ; is the sampling value With the kth distance neighbor The set of distance values between all elements in ; Get sample value The local reachable density is defined as: Calculate the sampling value based on the local reachable density obtained The local outlier factor is calculated as: In the formula, is the sampling value The local outlier factor of The sampled values are analyzed using the local outlier factor obtained The abnormal situation is judged according to the following principles: when >1, indicates sampling value It may be abnormal data, and when The larger the value, the greater the possibility of data anomaly; when Close to 1, indicates the sampled value It may belong to the same cluster as the neighboring points and is normal data; when <1, indicates sampling value are dense data points; The screened performance index parameters, dimmer brightness value, environmental parameters, and state of charge value are normalized to the interval [0, 1], and the normalization formula is: In the formula, represents the value after normalization; x represents the data value before normalization; , They are respectively represented as the maximum and minimum values of the data before normalization. The dimmer brightness value, environmental parameters, and charge state sampling values after normalization are used as input data, and the corresponding performance index parameter sampling values after normalization are used as output data. A BP neural network model is constructed, and the established neural network model is trained. The target error and maximum number of cycles of the network model are set. The number of hidden layers of the model is adjusted according to the error of the training result until the prediction accuracy is not lower than the expected value. The neural network model for predicting the performance index parameters is obtained. The calculation formula for the expected value of the prediction accuracy is, in, Expressed as the expected value of prediction accuracy; Represented as the input sample size of the neural network; Represented as the output function of the neural network; Represented as the kth output sample of the neural network; The constructed neural network model is used to obtain the predicted values of the power supply performance index parameters to be tested, and the correction coefficients of various performance index parameters are calculated based on the obtained predicted values. , the calculation formula is: In the formula, Represented as the input sample size of the neural network; It is expressed as the predicted value of the performance index parameter obtained by the neural network model, and the correction coefficient Correct the sampled values of the performance indicator parameters to obtain the corrected values of each performance indicator parameter. The calculation formula is: .
[0007] A laser power supply component performance detection system includes a dimmer, a power meter, a dynamic adjustment module, an environmental parameter acquisition module, a charge state acquisition module, a data processing module, a neural network construction module, and a data correction module; The dimmer is connected to the power supply to be detected, and is used to obtain the state of charge upper limit SOC of the power supply to be detected. max Continue to discharge to the lower limit of charge state SOC min The dimmer brightness sampling value during the process; The power meter is connected to the power supply to be detected, and is used to obtain the state of charge upper limit SOC of the power supply to be detected. maxContinue to discharge to the lower limit of charge state SOC min The process includes sampling values of performance index parameters such as rated power, actual power, and conversion efficiency; The dynamic adjustment module is used to adjust the state of charge SOC of the power supply to be detected. max Continue to discharge to the lower limit of charge state SOC min During the process, the brightness value of the dimmer is dynamically adjusted; The environmental parameter acquisition module is used to obtain the upper limit SOC of the state of charge of the power supply to be detected. max Continue to discharge to the lower limit of charge state SOC min The process includes sampling values of environmental parameters such as the operating temperature of the power supply to be tested and the humidity of the environment in which it is located; The state of charge acquisition module is used to acquire a sample value of the state of charge of the power supply to be detected; The data processing module is used to calculate the local outlier factor in the performance indicator parameter using the LOF algorithm, and use the obtained local outlier factor to calculate the jth sampling value of the i-th performance indicator. Determine whether it is abnormal data, filter and remove abnormal data in the performance indicator parameter sampling values; The neural network construction module is used to construct a BP neural network model with the normalized dimmer brightness value, environmental parameters, and charge state sampling values as input data, and the corresponding normalized performance index parameter sampling values as output data, and train the established neural network model, set the target error and maximum number of cycles of the network model, and adjust the number of hidden layers of the model according to the error of the training result until the prediction accuracy is not lower than the expected value, so as to obtain a neural network model for predicting the performance index parameter; The data correction module is used to obtain the predicted value of the power performance index parameter to be detected by using the constructed neural network model, and calculate the correction coefficient of each performance index parameter according to the obtained predicted value , the calculation formula is: , where Represented as the input sample size of the neural network; It is expressed as the predicted value of the performance index parameter obtained by the neural network model, and the correction coefficient Correct the sampled values of the performance indicator parameters to obtain the corrected values of each performance indicator parameter. The calculation formula is: ; The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor.
[0008] The present invention has the following beneficial effects: Compared with the prior art, the power supply to be detected is set to the upper limit of the state of charge SOCmax Continue to discharge to the lower limit of charge state SOC min The performance index parameters and the brightness value, environmental parameters, and state of charge value of the dimmer in the process are sampled, and the abnormal data in the performance index parameter sampling values are screened and eliminated for normalization. The normalized dimmer brightness value, environmental parameters, and state of charge sampling values are used as input data, and the corresponding normalized performance index parameter sampling values are used as output data. A BP neural network model is constructed and trained. The constructed neural network model is used to obtain the predicted values of the performance index parameters of the power supply to be detected, and the correction coefficients of various performance index parameters are calculated based on the obtained predicted values. The sampling values of the performance index parameters are corrected by the correction coefficient to obtain the corrected values of various performance index parameters. This scheme fully considers the influence of external factors such as environmental interference and battery status on the dynamic power performance measurement process of the laser power supply. By correcting the measurement results, the deviation between the measurement results and the actual situation can be reduced, and the accuracy of the measurement results can be improved, thereby truly reflecting the dynamic power performance of the laser power supply being tested. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 This is a flow chart of a method for detecting performance of a laser power supply assembly according to the present invention; Figure 2 This is a structural block diagram of a laser power supply component performance detection system of the present invention; Figure 3 This is a structural diagram of the neural network constructed in the solution of the present invention. DETAILED DESCRIPTION
[0010] The present invention will be further described below in conjunction with specific implementation methods, wherein the accompanying drawings are only used for exemplary descriptions and represent only schematic diagrams rather than actual drawings, and should not be understood as limiting the present invention. In order to better illustrate the specific implementation methods of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product.
[0011] The specific implementation process of the technical solution of the present invention includes the following steps: Step 1: Connect the power supply to be tested to the dimmer and power meter, dynamically adjust the brightness of the dimmer, and adjust the power supply to be tested from the upper limit of the state of charge SOC max Continue to discharge to the lower limit of charge state SOC min The performance index parameters of the process and the brightness value, environmental parameters and charge state value of the dimmer are sampled; Among them, the performance index parameters include rated power, actual power, and conversion efficiency; the environmental parameters include the operating temperature of the power supply to be tested and the humidity of the environment; the upper limit of the state of charge SOC maxThe value range is [0.8,0.9], and the lower limit of the state of charge SOC min The value range of is [0.15,0.2]; The specific steps for using a power meter to measure the rated power, actual power, and efficiency of a power supply are as follows: Check whether the power meter is intact and ensure that there is no damage, corrosion, looseness, etc. Ensure that the power supply voltage is stable and the power cord is firmly connected. Avoid installing the power meter in an environment affected by vibration, moisture and high temperature to ensure stable operation of the equipment; According to the characteristics of the circuit and the requirements of the power meter, select the appropriate connection method, such as series or parallel connection, and correctly connect the circuit to be tested to the input terminal of the power meter to ensure that the connection is firm and the contact is good; Before using the power meter, it must be calibrated to ensure accuracy and set measurement parameters such as power range, sampling rate, display format, etc. according to actual needs; Stabilize circuit operation, avoid interference and fluctuations, and select measurement functions; Read the measurement results and pay attention to whether the display is stable to avoid measurement errors caused by jitter or instability; Record the monitored data and calculate the rated power and actual power of the power supply, as well as the efficiency.
[0012] The calculation formula for power efficiency is: η=P out / P in ×100%, where P out is the output power of the power supply, P in It is the input power of the power supply. During the measurement process, the output voltage and current can be directly measured using a power meter to obtain the output power, while the input voltage and current can be measured to obtain the input power. Step 2: Use the LOF algorithm to calculate the local outlier factor in the performance indicator parameters, and use the obtained local outlier factor to calculate the jth sampling value of the i-th performance indicator. Whether it is abnormal data is determined, and abnormal data in the performance indicator parameter sampling values is screened and removed. The determination process includes the following steps: Step 21: Calculate the sample values separately With the rest of the sample values The distance between The calculation formula is: Among them, j, q∈N, and q≠j, N is the total amount of sampled data; Step 22: Get the distance value Sort by small to large, set and sample value The kth element closest to p∈N, and p≠j, get the sample value With the sample value The kth distance value in, ; Step 23: According to the kth distance value Identify the elements The kth distance neighbor of ; Step 24: Get the sample value The k-th reachable distance of , is the sampling value The k-th reachable distance of To obtain , The maximum value operation in ; is the sampling value With the kth distance neighbor The set of distance values between all elements in ; Step 25: Get the sample value The local reachable density is defined as: ; Step 26: Calculate the sampling value based on the obtained local reachability density The local outlier factor is calculated as: In the formula, is the sampling value The local outlier factor of Step 27: Use the obtained local outlier factor to analyze the sampled values The abnormal situation is judged according to the following principles: when >1, indicates sampling value It may be abnormal data, and when The larger the value, the greater the possibility of data anomaly; when Close to 1, indicates the sampled value It may belong to the same cluster as the neighboring points and is normal data; when <1, indicates sampling value are dense data points; Step 3: Normalize the selected performance index parameters and dimmer brightness value, environmental parameters, and state of charge value to the [0, 1] interval. The normalization formula is: In the formula, represents the value after normalization; x represents the data value before normalization; , They are respectively represented as the maximum and minimum values of the data before normalization. The dimmer brightness value, environmental parameters, and charge state sampling values after normalization are used as input data, and the corresponding performance index parameter sampling values after normalization are used as output data. A BP neural network model is constructed, and the established neural network model is trained. The target error and maximum number of cycles of the network model are set. The number of hidden layers of the model is adjusted according to the error of the training result until the prediction accuracy is not lower than the expected value. The neural network model for predicting the performance index parameters is obtained. The calculation formula for the expected value of the prediction accuracy is, in, Expressed as the expected value of prediction accuracy; Represented as the input sample size of the neural network; Represented as the output function of the neural network; Represented as the kth output sample of the neural network; For the BP neural network model in this scheme, the input layer includes the dimmer brightness value, the operating temperature of the power supply to be detected, the humidity of the environment, and the state of charge. The number of neuron nodes in the input layer is 4. The output layer includes the rated power, actual power, and conversion efficiency. The number of neuron nodes in the output layer is 3. The number of neuron nodes in the intermediate layer can be determined according to the empirical formula, which is: Where a is a constant from 1 to 9. Expressed as a pair The calculation result is rounded up, so that the structure of the neural network can be determined as an input layer of 4 neuron nodes, an intermediate layer of s neuron nodes, and an output layer of 3 neuron nodes, as shown in Figure 3 As shown; In this embodiment, taking the use of MATLAB software to construct a neural network model as an example, the network is trained using a standard BP neural network algorithm, the target error and the maximum number of cycles are set, the transfer function of the hidden layer neurons is selected to be the S-type tangent function tansig, the transfer function of the output layer neurons is selected to be the S-type logarithmic function logsig, the traingdx function is selected as the training function, and the number of intermediate layer neuron nodes is continuously adjusted so that the prediction accuracy is not lower than the expected value, and the neural network model for predicting performance index parameters can be obtained; Step 4: Use the constructed neural network model to obtain the predicted values of the power supply performance index parameters to be tested, and calculate the correction coefficients of various performance index parameters based on the obtained predicted values , the calculation formula is: In the formula, Represented as the input sample size of the neural network; It is expressed as the predicted value of the performance index parameter obtained by the neural network model, and the correction coefficient Correct the sampled values of the performance indicator parameters to obtain the corrected values of each performance indicator parameter. The calculation formula is: .
[0013] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A method for detecting the performance of a laser power supply assembly, characterized in that: include: Connect the power supply to be tested to the dimmer and power meter, dynamically adjust the brightness of the dimmer, and set the power supply to be tested from the upper limit of the state of charge SOC max Continue to discharge to the lower limit of charge state SOC min The performance index parameters of the process and the brightness value, environmental parameters and charge state value of the dimmer are sampled; The LOF algorithm is used to calculate the local outlier factor in the performance indicator parameter, and the jth sampling value of the i-th performance indicator is calculated using the obtained local outlier factor. Determine whether it is abnormal data, filter and remove abnormal data in the performance indicator parameter sampling values; The screened performance index parameters and dimmer brightness value, environmental parameters, and state of charge value are normalized to the interval [0, 1], the normalized dimmer brightness value, environmental parameters, and state of charge sampling values are used as input data, and the corresponding normalized performance index parameter sampling values are used as output data, a BP neural network model is constructed, the established neural network model is trained, the target error and maximum number of cycles of the network model are set, and the number of hidden layers of the model is adjusted according to the error of the training result until the prediction accuracy is not lower than the expected value, thereby obtaining a neural network model for predicting the performance index parameters; The constructed neural network model is used to obtain the predicted values of the power supply performance index parameters to be tested, and the correction coefficients of various performance index parameters are calculated based on the obtained predicted values. The calculation formula is: , where Represented as the input sample size of the neural network; It is expressed as the predicted value of the performance index parameter obtained by the neural network model, and the correction coefficient Correct the sampled values of the performance indicator parameters to obtain the corrected values of each performance indicator parameter. The calculation formula is: .
2. A method for detecting performance of a laser power supply assembly according to claim 1, characterized in that: The abnormal data determination process in the performance indicator parameter sampling value includes the following steps: Calculate the sample values separately With the rest of the sample values The distance between , the calculation formula is: , where j, q∈N, and q≠j, and N is the total amount of sampled data; The distance value to be obtained Sort by small to large, set and sample value The kth element closest to , p∈N, and p≠j, get the sample value With the sample value The kth distance value ,in, ; According to the kth distance value Identify the elements The kth distance neighbor of ; Get sample value The k-th reachable distance of , is the sampling value The k-th reachable distance of To obtain , The maximum value operation in ; is the sampling value With the kth distance neighbor The set of distance values between all elements in ; Get sample value The local reachable density is defined as: ; Calculate the sampling value based on the local reachable density obtained The local outlier factor is calculated as: In the formula, is the sampling value The local outlier factor of The sampled values are analyzed using the local outlier factor obtained The abnormal situation is judged according to the following principles: when >1, indicates sampling value It may be abnormal data, and when The larger the value, the greater the possibility of data anomaly; when Close to 1, indicates the sample value It may belong to the same cluster as the neighboring points and is normal data; when <1, indicates sampling value are dense data points.
3. A method for detecting performance of a laser power supply assembly according to claim 1, characterized in that: The performance index parameters include rated power, actual power, and conversion efficiency; The environmental parameters include the operating temperature of the power supply to be detected and the humidity of the environment in which it is located.
4. A method for detecting performance of a laser power supply assembly according to claim 1, characterized in that: The data normalization formula is: In the formula, represents the value after normalization; x represents the data value before normalization; , They are respectively represented as the maximum and minimum values in the data before normalization.
5. A method for detecting performance of a laser power supply assembly according to claim 1, characterized in that: The calculation formula for the expected value of prediction accuracy is: in, Expressed as the expected value of prediction accuracy; Represented as the input sample size of the neural network; Represented as the output function of the neural network; Represented as the kth output sample of the neural network.
6. A method for detecting performance of a laser power supply assembly according to claim 1, characterized in that: State of charge upper limit SOC max The value range is [0.8,0.9], and the lower limit of the state of charge SOC min The value range is [0.15,0.2].
7. A laser power supply component performance detection system, characterized in that: It includes a dimmer, a power meter, a dynamic adjustment module, an environmental parameter acquisition module, a charge state acquisition module, a data processing module, a neural network construction module, and a data correction module; The dimmer is connected to the power supply to be detected, and is used to obtain the state of charge upper limit SOC of the power supply to be detected. max Continue to discharge to the lower limit of charge state SOC min The dimmer brightness sampling value during the process; The power meter is connected to the power supply to be detected, and is used to obtain the state of charge upper limit SOC of the power supply to be detected. max Continue to discharge to the lower limit of charge state SOC min The process includes sampling values of performance index parameters such as rated power, actual power, and conversion efficiency; The dynamic adjustment module is used to adjust the state of charge SOC of the power supply to be detected. max Continue to discharge to the lower limit of charge state SOC min During the process, the brightness value of the dimmer is dynamically adjusted; The environmental parameter acquisition module is used to obtain the upper limit SOC of the state of charge of the power supply to be detected. max Continue to discharge to the lower limit of charge state SOC min The process includes the environmental parameter sampling values of the operating temperature of the power supply to be tested and the humidity of the environment in which it is located, among which the upper limit of the state of charge SOC max The value range is [0.8,0.9], and the lower limit of the state of charge SOC min The value range of is [0.15,0.2]; The state of charge acquisition module is used to acquire a sample value of the state of charge of the power supply to be detected; The data processing module is used to calculate the local outlier factor in the performance indicator parameter using the LOF algorithm, and use the obtained local outlier factor to calculate the jth sampling value of the i-th performance indicator. Whether it is abnormal data is determined, and abnormal data in the performance indicator parameter sampling values is screened and removed. The determination process includes the following steps: Calculate the sample values separately With the rest of the sample values The distance between , the calculation formula is: , where j, q∈N, and q≠j, and N is the total amount of sampled data; The distance value to be obtained Sort by small to large, set and sample value The kth element closest to p∈N, and p≠j, get the sample value With the sample value The kth distance value ,in, ; According to the kth distance value Identify the elements The kth distance neighbor of ; Get sample value The k-th reachable distance of , is the sampling value The k-th reachable distance of To obtain , The maximum value operation in ; is the sampling value With the kth distance neighbor A collection of distance values between all elements in; Get the sampled value The local reachable density is defined as: ; Calculate the sampling value based on the local reachable density obtained The local outlier factor is calculated as: In the formula, is the sampling value The local outlier factor of The sampled values are analyzed using the local outlier factor obtained The abnormal situation is judged according to the following principles: when >1, indicates sampling value It may be abnormal data, and when The larger the value, the greater the possibility of data anomaly; when Close to 1, indicates the sample value It may belong to the same cluster as the neighboring points and is normal data; when <1, indicates sampling value are dense data points; The neural network construction module is used to construct a BP neural network model with the normalized dimmer brightness value, environmental parameters, and charge state sampling values as input data, and the corresponding normalized performance index parameter sampling values as output data, and train the established neural network model, set the target error and maximum number of cycles of the network model, and adjust the number of hidden layers of the model according to the error of the training result until the prediction accuracy is not lower than the expected value, so as to obtain a neural network model for predicting the performance index parameter; The data correction module is used to obtain the predicted value of the power performance index parameter to be detected by using the constructed neural network model, and calculate the correction coefficient of each performance index parameter according to the obtained predicted value , the calculation formula is: , where Represented as the input sample size of the neural network; It is expressed as the predicted value of the performance index parameter obtained by the neural network model, and the correction coefficient Correct the sampled values of the performance indicator parameters to obtain the corrected values of each performance indicator parameter. The calculation formula is: .
8. A laser power supply component performance detection system according to claim 7, characterized in that: The system comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the steps of the method according to any one of claims 1 to 6 are implemented when the processor executes the program.
Citation Information
Patent Citations
Laser power supply testing system and method
CN109031158A
Life prediction method and system of laser power supply, and storage medium
CN116882303A
Laser power supply aging test method and system
CN118011265A
Cited By
Water pump operation energy-saving control method and system based on multi-source data analysis
CN119686974A
