Communication abnormity identification method for numerical control power supply

Abstract

The invention discloses a communication abnormity identification method of a numerical control power supply, and relates to the field of numerical control power supplies, and the method comprises the following steps: 1, deploying data collection units at power transmission equipment and communication equipment, and constructing an abnormity type database; 2, constructing a response simulation model, receiving a to-be-issued control instruction by the response simulation model, and simulating and outputting simulation response data which is provided by the sensing equipment, the power supply equipment and the communication equipment and comprises a plurality of corresponding items; 3, after the control instruction is issued, collecting actual response data including a plurality of corresponding items of the sensing equipment, the power supply equipment and the communication equipment; according to the method, data quantification processing is carried out on communication abnormity problems in real time, abnormal data are converted into an analyzable form, so that fault types are identified more accurately, the abnormity processing process is optimized, the abnormal data are classified effectively according to the characteristics and types of the abnormal data through reasonable abnormal data classification, and abnormal conditions are better understood and analyzed.

Description

Technical Field

[0001] The present invention relates to the technical field of digital control power supplies, and in particular to a method for identifying communication anomalies of a digital control power supply. Background Art

[0002] The communication principle of CNC power supply mainly involves the information exchange between power supply equipment and control system to realize the monitoring and adjustment of power supply parameters. During the communication process of CNC power supply, abnormal faults may affect the normal operation of equipment and the reliability of the system. In the actual operation process, communication abnormalities may cause data loss or data errors, thus affecting the normal operation of the entire system and may also cause damage to the entire system.

[0003] However, the existing technology for identifying communication anomalies still has some deficiencies, such as:

[0004] 1. CNC power supply systems usually include multiple devices and platforms, such as communication equipment, sensor equipment and power supply equipment. Due to factors such as signal interference or equipment failure, various abnormal problems may occur during the communication process. When abnormal problems occur, it is difficult to quantify the data of the abnormal problems in time, and it is difficult to accurately identify the fault type, which in turn affects the arrangement of reasonable processing behaviors;

[0005] 2. When abnormal data is initially identified, it is difficult to effectively classify the abnormal data. When there is a large amount of real-time data, when anomalies occur, the abnormal data is also complex and huge, and the abnormal types are not unique, making it difficult to accurately and effectively trace the abnormal data. Summary of the invention

[0006] 1. Technical issues to be solved

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a method for identifying communication anomalies of a digital control power supply, which can effectively solve the problems of the prior art.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] The present invention discloses a method for identifying communication anomalies of a digital control power supply, comprising the following steps:

[0011] Step 1: Deploy data collection units at power transmission equipment and communication equipment to build an abnormal type database;

[0012] Step 2: Construct a response simulation model, the response simulation model receives the control instructions to be issued, and simulates and outputs the simulation response data provided by each sensor device, power supply device and communication device, including corresponding items;

[0013] Step 3: After the control command is issued, actual response data including several corresponding items of each sensor device, power supply device and communication device are collected;

[0014] Step 4: Determine whether the actual response data obtained is consistent with the expected response;

[0015] Step 5: If the judgment is yes, then continue to control the instruction sending behavior;

[0016] Step 6: If the judgment is no, then the actual response data is obtained and compared with the simulated response data, and the corresponding deviation values ​​of several reference targets are calculated;

[0017] Step 7: Based on each abnormal type in the abnormal type database, analyze the associated target influence coefficients of several corresponding deviation values ​​currently obtained;

[0018] Step 8: Based on the associated target impact coefficient, classify the deviation data into the corresponding abnormality type, output the abnormality type, the source, the severity based on a certain source, and the equipment or procedures that can be adjusted;

[0019] Step 9: Record the classification results, generate anomaly identification report and submit it to the management end.

[0020] Furthermore, the construction and operation process of the response simulation model in step 2 is as follows:

[0021] Step 201: Collect historical actual response data of the sensor device, power supply device and communication device and setting data when responding;

[0022] Step 202: Clean and standardize the obtained historical data and divide them into a training set and a validation set;

[0023] Step 203: Select a multi-layer perceptron neural network structure to build a response simulation model and initialize the network weights;

[0024] Step 204: training the response simulation model using the training set, and adjusting weights and biases through a back propagation algorithm;

[0025] Step 205: Use the validation set data to evaluate the model performance and adjust the model parameters according to the validation results;

[0026] Step 206: deploy the trained response simulation model to the communication system, receive control instructions in real time and generate simulation response data.

[0027] Furthermore, the several types of items corresponding to the simulated response data in step 2 include: data transmission rate, signal strength, bit rate, delay time, packet loss rate and bandwidth utilization in the communication equipment items corresponding to each power transmission node, temperature sensor readings, humidity sensor readings, pressure sensor readings, vibration sensor readings, current sensor readings, voltage sensor readings and flow sensor readings in the sensor equipment items, and charging response efficiency, cooling component response efficiency, power factor, harmonic distortion rate and remaining capacity percentage in the power supply equipment items.

[0028] Furthermore, the exception type database in step 1 obtains historical actual response data, simulated response data and the situation when the exception occurs, removes abnormal data, processes missing values ​​and handles abnormal values, and then performs comprehensive extraction to obtain the exception type description, target parameters affected by the exception, the cause of the exception and historical solution data.

[0029] Furthermore, in the process of calculating the corresponding deviation values ​​of several reference targets, step 6 obtains actual response data and simulated response data, obtains the reference targets that need to be compared determined in step 4, and calculates the deviation value of each reference target.

[0030] Furthermore, when analyzing the associated target impact coefficient in step 7, the abnormal type processing priority is evaluated, and the linear correlation between each abnormal type and the corresponding deviation value is calculated by the Pearson correlation coefficient according to the evaluation priority. The calculation formula is:

[0031]

[0032] In the formula, n represents the number of samples, x i represents the value of the variable x of the i-th sample, y i represents the value of the variable y of the i-th sample, represents the mean value of variable x, Represents the mean value of the variable y.

[0033] Furthermore, in the evaluation process of the exception type processing priority, the impact target, the impact size and frequency factors of each exception type are quantified, and a comprehensive evaluation is performed based on the quantified results. An importance weight is determined for each exception type, and the weighted sum method is used for calculation to obtain the priority order of each exception type.

[0034] Furthermore, in the process of classifying the deviation data in step 8 into the corresponding abnormal type, if the deviation data involved in the classification cannot be successfully classified, the characteristics, change trends and related abnormal sources of the data are checked, and the inspection data is obtained for manual inspection and program inspection. According to the results of manual and program inspection, it is determined whether to define a new abnormal type. If the definition is successful, the abnormal type database is updated.

[0035] Furthermore, in step 8, based on the classification results of the deviation data, the associated types are matched in the abnormal type database, the actual response data is analyzed, and traced back to a specific one or more sensor devices, power supply equipment or communication equipment. According to the impact coefficient of the abnormality and the importance coefficient to the system operation, the severity of the abnormality from a certain source is evaluated, and according to the type of abnormality and the source of generation, the equipment or program that can be involved in the adjustment is determined.

[0036] Furthermore, the abnormal report is uploaded to the cloud for backup, supporting writing and reading of external storage media.

[0037] (III) Beneficial effects

[0038] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects:

[0039] 1. By quantifying the data of communication anomalies in real time and converting the abnormal data into an analyzable form, the fault type can be identified more accurately, thereby optimizing the abnormal handling process and improving the processing efficiency. By building a response simulation model, obtaining simulated response data, and calculating the deviation between the actual response data and the simulated response data, the abnormal state can be accurately identified.

[0040] 2. Through reasonable classification of abnormal data, the abnormal data can be effectively classified according to its characteristics and types, so as to better understand and analyze the abnormal situation. The abnormal data can be traced, that is, the source and path of the abnormal data can be determined, and then the source of the abnormality can be quickly located, so as to facilitate the subsequent adoption of corresponding measures for processing, and facilitate managers to deeply understand the causes of the abnormality and take corresponding corrective measures, thereby improving the stability and reliability of the system.

[0041] 3. By covering the entire process from data collection, simulation prediction to anomaly identification, automatic anomaly detection is achieved through model prediction and data comparison. Accurate analysis is performed using deviation values ​​and influence coefficients. It not only identifies anomalies, but also provides practical information on anomaly type, source and severity, effectively improving the reliability and maintenance efficiency of CNC power supply systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] The present invention will be further described below in conjunction with the embodiments.

[0046] Example 1

[0047] A communication abnormality identification method of a digital control power supply in this embodiment, such as Figure 1 As shown, the following steps are included:

[0048] Step 1: Deploy data acquisition units at power transmission equipment and communication equipment to build an abnormality type database. The abnormality type database obtains historical actual response data, simulated response data, and the situation when the abnormality occurs, removes abnormal data, processes missing values, and processes abnormal values ​​before comprehensive extraction. It obtains abnormality type descriptions, target parameters affected by abnormalities, abnormal causes, and historical solution data to provide basic data and reference standards for subsequent analysis.

[0049] Step 2: Construct a response simulation model. The response simulation model receives the control instructions to be issued, and simulates and outputs the simulated response data provided by each sensor device, power supply device and communication device, which includes several corresponding items. The types of items corresponding to the simulated response data include: data transmission rate, signal strength, bit rate, delay time, packet loss rate and bandwidth utilization rate in the communication equipment project corresponding to each power transmission node, temperature sensor reading, humidity sensor reading, pressure sensor reading, vibration sensor reading, current sensor reading, voltage sensor reading and flow sensor reading in the sensor equipment project, charging response efficiency, cooling component response efficiency, power factor, harmonic distortion rate and remaining capacity percentage in the power supply equipment project; the response simulation model can be used to predict the system response under normal conditions and provide a benchmark for abnormality detection;

[0050] The construction and operation process of the response simulation model is as follows:

[0051] Step 201: Collect historical actual response data of the sensor device, power supply device and communication device and setting data when responding;

[0052] Step 202: Clean and standardize the obtained historical data and divide them into a training set and a validation set;

[0053] Step 203: Select a multi-layer perceptron neural network structure to build a response simulation model and initialize the network weights. This neural network structure has strong adaptive and generalization capabilities and can adapt to different abnormal situations and data distributions.

[0054] Step 204: training the response simulation model using the training set, and adjusting weights and biases through a back propagation algorithm;

[0055] Step 205: Use the validation set data to evaluate the model performance and adjust the model parameters according to the validation results;

[0056] Step 206: deploying the trained response simulation model to the communication system, receiving control instructions in real time and generating simulation response data;

[0057] Step 3: After the control command is issued, the actual response data of each sensor device, power supply device and communication device, including several corresponding items, is collected to obtain the real operation data for comparison with the simulation data;

[0058] Step 4: Determine whether the actual response data obtained meets the expected response, quickly screen out situations where there may be problems, and improve efficiency;

[0059] Step 5: If the judgment is yes, then continue to control the instruction sending behavior;

[0060] Step 6: If the judgment is no, then the actual response data is obtained for comparison with the simulated response data, and the corresponding deviation values ​​of several reference targets are calculated. In the process of calculating the corresponding deviation values ​​of several reference targets, the actual response data and the simulated response data are obtained, and the reference targets to be compared obtained in step 4 are obtained, and the deviation value of each reference target is calculated. The specific deviation is found through quantitative analysis to provide a basis for subsequent analysis;

[0061] Step 7: Based on each abnormal type in the abnormal type database, analyze the associated target impact coefficients of the corresponding deviation values ​​currently obtained, use the pre-established database to perform intelligent abnormal diagnosis and classification, use the abnormal type database to compare the calculated deviation values, analyze the association between each deviation value and the known abnormal type, and determine the impact of each abnormality on the system;

[0062] Step 8: Based on the associated target impact coefficient, classify the deviation data into the corresponding abnormality type, output the abnormality type, the source, the severity based on a certain source, and the equipment or procedures that can be adjusted;

[0063] Step 9: Record the classification results, generate an exception identification report and submit it to the management end. The exception report is uploaded to the cloud for backup, supports writing and reading of external storage media, and provides detailed exception information. Through precise simulation and comparison mechanisms, it can effectively identify and classify communication anomalies in the CNC power supply system, ensure the stable operation of the system, and facilitate subsequent exception processing and analysis.

[0064] Compared with the existing technology, by deploying data acquisition units at power transmission equipment and communication equipment, it is possible to comprehensively obtain historical actual response data, simulated response data and situations when abnormalities occur, thereby building a complete abnormality type database to improve the accuracy and reliability of abnormality identification, and using the response simulation model to simulate the output of the control instructions to be issued, generate simulated response data of each sensor device, power supply equipment and communication equipment, and simulate and verify the system before actual operation, reducing the risk and cost of actual operation;

[0065] A multilayer perceptron neural network structure is selected to construct a response simulation model. By training and adjusting the model parameters, the actual response data can be effectively simulated and predicted. By comparing the actual response data with the simulated response data, the deviation value of the reference target is calculated, and the associated target influence coefficient in the abnormal type database is analyzed to classify the deviation data into the corresponding abnormal type.

[0066] Example 2

[0067] At other levels, this embodiment also provides a method for evaluating the priority of abnormal type processing. When analyzing the associated target impact coefficient, the abnormal type processing priority is evaluated. During the evaluation of the abnormal type processing priority, the impact target, the impact size and frequency factors of each abnormal type are quantified. A comprehensive evaluation is performed based on the quantified results, and an importance weight is determined for each abnormal type. The weighted sum method is used for calculation to obtain the priority order of each abnormal type.

[0068] According to the evaluation priority, the linear correlation between each abnormal type and the corresponding deviation value is calculated by the Pearson correlation coefficient. The calculation formula is:

[0069]

[0070] In the formula, n represents the number of samples, x i represents the value of the variable x of the i-th sample, y i represents the value of the variable y of the i-th sample, represents the mean value of variable x, It represents the average value of variable y, the numerator is the covariance of two variables x and y, and the denominator is the product of the standard deviations of two variables x and y. The correlation coefficient ranges between -1 and 1. When the correlation coefficient is 1, it means that the two variables are completely positively correlated. When the correlation coefficient is -1, it means that the two variables are completely negatively correlated. When the correlation coefficient is 0, it means that the two variables have no linear relationship. Through correlation analysis, we can get the degree of linear correlation between two variables, which helps us understand the relationship and trend between them, as well as the impact on one of the variables.

[0071] Example 3

[0072] In this embodiment, during the process of classifying the deviation data into the corresponding abnormal type, if the deviation data involved in the classification cannot be successfully classified, the characteristics, change trends and associated abnormal sources of the data are checked, and the inspection data is obtained for manual inspection and program inspection. According to the results of manual and program inspection, it is determined whether to define a new abnormal type. If the definition is successful, the abnormal type database is updated. The abnormal type database can be updated in time to improve the flexibility and adaptability of abnormal identification.

[0073] According to the classification results of the deviation data, the associated types are matched in the abnormal type database, the actual response data is analyzed, and traced back to a specific one or more sensor devices, power supply equipment or communication equipment. According to the impact coefficient of the abnormality and the importance coefficient to the system operation, the severity of the abnormality from a certain source is evaluated. According to the type of abnormality and the source of generation, the equipment or program that can be involved in the adjustment is determined, which helps to take corresponding adjustment measures in a targeted manner and improve the stability and reliability of the system.

[0074] Compared with the existing technology, the severity of a certain abnormal source can be evaluated according to the impact coefficient of the abnormality and the importance coefficient to the system operation, and the abnormal situations that have a greater impact on the system can be handled with priority, which can improve the accuracy, flexibility and operability of abnormality identification and processing.

[0075] In summary, when the present invention is applied, the operator deploys the data acquisition unit at the power transmission equipment and the communication equipment, collects the actual response data and setting data of each sensor device, power supply device and communication device in real time, and stores them in a database, uses the historical actual response data and related setting data to train and evaluate the response simulation model, and after the control instruction is issued, collects the actual response data of each sensor device, power supply device and communication device including several corresponding items;

[0076] Then, these actual response data are compared with the simulated response data. During the comparison process, it is necessary to determine whether the deviation value of each reference target meets the expected value. If the actual response data meets the expected value, the instruction sending behavior is continuously controlled. If the corresponding deviation value between the actual response data and the simulated response data exceeds the preset threshold, it is determined to be an abnormal situation. According to the associated target influence coefficient of each abnormal type in the abnormal type database, the deviation data can be classified into the corresponding abnormal type, and the type of abnormality, the source of generation, the severity based on a certain source, the equipment or program that can be adjusted, and other information are output;

[0077] The present invention combines actual response data and simulated response data for comparison, can more accurately identify abnormal situations, improve recognition accuracy, and by building a response simulation model, can simulate various abnormal situations in the actual system, realize online or offline diagnosis, improve processing efficiency, and based on the abnormal type database, can quickly identify and classify abnormal types, thereby improving the efficiency of handling abnormalities;

[0078] By quantifying the data of communication anomalies in real time and converting the anomaly data into an analyzable form, the fault type can be identified more accurately, and the anomaly handling process can be optimized to improve the handling efficiency. By building a response simulation model, obtaining simulated response data, and calculating the deviation between the actual response data and the simulated response data, the abnormal state can be accurately identified.

[0079] In the assessment of the priority of abnormal type processing, the impact target, the magnitude of the impact and the frequency factors of the abnormal type are comprehensively considered, which can more objectively assess the importance of the abnormal situation. By uploading the abnormal report to the cloud backup and writing and reading the external storage media, data sharing and backup management are facilitated;

[0080] By reasonably classifying abnormal data, the abnormal data can be effectively classified according to its characteristics and types, so as to better understand and analyze abnormal situations. The abnormal data can be traced, that is, the source and path of the abnormal data can be determined, and then the source of the abnormality can be quickly located, so as to facilitate the subsequent adoption of corresponding measures for processing, and facilitate managers to deeply understand the causes of the abnormality and take corresponding corrective measures, thereby improving the stability and reliability of the system.

[0081] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying communication anomalies of a digital control power supply, It is characterized in that The following steps are involved: Step 1: Deploy data collection units at power transmission equipment and communication equipment to build an abnormal type database; Step 2: Construct a response simulation model, the response simulation model receives the control instructions to be issued, and simulates and outputs the simulation response data provided by each sensor device, power supply device and communication device, including corresponding items; Step 3: After the control command is issued, actual response data including several corresponding items of each sensor device, power supply device and communication device are collected; Step 4: Determine whether the actual response data obtained is consistent with the expected response; Step 5: If the judgment is yes, then continue to control the instruction sending behavior; Step 6: If the judgment is no, then the actual response data is obtained and compared with the simulated response data, and the corresponding deviation values ​​of several reference targets are calculated; Step 7: Based on each abnormal type in the abnormal type database, analyze the associated target influence coefficients of several corresponding deviation values ​​currently obtained; Step 8: Based on the associated target impact coefficient, classify the deviation data into the corresponding abnormality type, output the abnormality type, the source, the severity based on a certain source, and the equipment or procedures that can be adjusted; Step 9: Record the classification results, generate anomaly identification report and submit it to the management end.

2. A method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that The construction and operation process of the response simulation model in step 2 is as follows: Step 201: Collect historical actual response data of the sensor device, power supply device and communication device and setting data when responding; Step 202: Clean and standardize the obtained historical data and divide them into a training set and a validation set; Step 203: Select a multi-layer perceptron neural network structure to build a response simulation model and initialize the network weights; Step 204: training the response simulation model using the training set, and adjusting weights and biases through a back propagation algorithm; Step 205: Use the validation set data to evaluate the model performance and adjust the model parameters according to the validation results; Step 206: deploy the trained response simulation model to the communication system, receive control instructions in real time and generate simulation response data.

3. The method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that The several types of items corresponding to the simulated response data in the step 2 include: data transmission rate, signal strength, bit rate, delay time, packet loss rate and bandwidth utilization in the communication equipment items corresponding to each power transmission node, temperature sensor readings, humidity sensor readings, pressure sensor readings, vibration sensor readings, current sensor readings, voltage sensor readings and flow sensor readings in the sensor equipment items, and charging response efficiency, cooling component response efficiency, power factor, harmonic distortion rate and remaining capacity percentage in the power supply equipment items.

4. The method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that The abnormal type database in step 1 obtains historical actual response data, simulated response data and the situation when the abnormality occurs, removes abnormal data, processes missing values ​​and handles abnormal values, and then performs comprehensive extraction to obtain the abnormal type description, target parameters affected by the abnormality, abnormal causes and historical solution data.

5. The method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that In the process of calculating the corresponding deviation values ​​of several reference targets, step 6 obtains the actual response data and the simulated response data, obtains the reference targets that need to be compared determined in step 4, and calculates the deviation value of each reference target.

6. A method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that When analyzing the associated target impact coefficient in step 7, the abnormal type processing priority is evaluated. According to the evaluation priority, the linear correlation between each abnormal type and the corresponding deviation value is calculated by the Pearson correlation coefficient, and the calculation formula is: In the formula, n represents the number of samples, x i represents the value of the variable x of the i-th sample, y i represents the value of the variable y of the i-th sample, represents the mean value of variable x, Represents the mean value of the variable y.

7. A method for identifying communication anomalies of a digital control power supply according to claim 6, It is characterized in that During the evaluation process of the exception type processing priority, the impact target, the impact size and frequency factors of each exception type are quantified, and a comprehensive evaluation is performed based on the quantified results. An importance weight is determined for each exception type, and the priority order of each exception type is obtained by calculation through the weighted sum method.

8. The method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that In the process of classifying the deviation data into the corresponding abnormal type in step 8, if the deviation data involved in the classification cannot be successfully classified, the characteristics, change trends and related abnormal sources of the data are checked, and the inspection data is obtained for manual inspection and program inspection. According to the results of manual and program inspection, it is determined whether to define a new abnormal type. If the definition is successful, the abnormal type database is updated.

9. The method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that In step 8, based on the classification results of the deviation data, the associated types are matched in the abnormal type database, the actual response data is analyzed, and traced back to a specific one or more sensor devices, power supply equipment or communication equipment. According to the impact coefficient of the abnormality and the importance coefficient to the system operation, the severity of the abnormality from a certain source is evaluated, and according to the type of abnormality and the source of generation, the equipment or program that can be involved in the adjustment is determined.

10. A method for identifying communication anomalies of a digital control power supply according to claim 1, It is characterized in that The abnormal report is uploaded to the cloud for backup, supporting writing and reading of external storage media.

Images