HPLC fault diagnosis method

Abstract

The invention relates to an HPLC (High Performance Liquid Chromatography) fault diagnosis method, which relates to the field of HPLC fault identification, and comprises the following steps: responding to a fault signal, and obtaining historical data information corresponding to the fault signal; inputting the historical data information into a fault analysis model to obtain a fault probability corresponding to the fault reason; outputting an error fault signal when all the fault probabilities do not exceed a fault probability threshold; when the fault probability exceeds a fault probability threshold, defining the corresponding fault probability as a suspected fault probability, and defining a fault reason corresponding to the suspected fault probability as a suspected fault reason; obtaining a fault analysis scheme from a fault analysis scheme library based on the suspected fault reason; executing the fault analysis scheme to obtain a credibility parameter; and when the credibility parameter is higher than the credibility threshold, outputting the suspected fault reason corresponding to the credibility parameter. According to the method, the diagnosis efficiency and scientificity are improved through a full-automatic process, the reliability of a diagnosis result is ensured by depending on a bidirectional verification mechanism, and an efficient and accurate fault diagnosis effect is realized.

Description

Technical Field

[0001] The present invention relates to the field of HPLC fault identification, in particular to a HPLC fault diagnosis method. Background Art

[0002] HPLC (High-Speed ​​Power Line Carrier Communication), a data transmission method utilizing power lines, significantly improves data transmission efficiency compared to traditional power line communication due to its high transmission rate, strong anti-interference capabilities, and wide coverage. However, HPLC communication units can experience data transmission anomalies during operation due to factors such as complex power line environments and line aging. Therefore, timely HPLC fault diagnosis is essential to quickly restore communication, minimize financial losses, and provide a basis for preventive maintenance, thereby reducing operational costs.

[0003] Currently, HPLC fault diagnosis relies primarily on on-site manual investigations. Technicians must carry specialized equipment, such as signal analyzers and multimeters, to the faulty area. They conduct on-site tests of terminal communication quality, check line connections, verify device configuration parameters, and analyze key indicators such as RSSI, SNR, and BER. Comparing normal and abnormal data, technicians locate the fault point and rely on experience to determine if there is noise interference, line aging, or equipment compatibility issues.

[0004] Regarding the above-mentioned related technologies, technicians can only analyze faults based on the current limited data, which makes the fault analysis rely on the technicians' personal level and lacks scientific fault judgment, resulting in the HPLC may not be maintained in a timely and accurate manner. Summary of the Invention

[0005] In order to timely and scientifically diagnose HPLC faults and reduce the influence of technicians' personal level on HPLC fault repair, the present invention provides an HPLC fault diagnosis method.

[0006] The present invention provides a HPLC fault diagnosis method, which adopts the following technical solution: A HPLC fault diagnosis method, characterized by comprising: Step 1: In response to a fault signal, obtaining historical data information on a line corresponding to the fault signal; Step 2: Inputting the historical data information into a preset fault analysis model to obtain the failure probabilities corresponding to all preset fault causes; Step 20: Outputting a preset error fault signal when all the fault probabilities do not exceed a preset fault probability threshold; Step 21: When the failure probability exceeds the failure probability threshold, the corresponding failure probability is defined as a suspected failure probability, and the failure cause corresponding to the suspected failure probability is defined as a suspected failure cause; Step 210: Obtaining a fault analysis solution from a preset fault analysis solution library based on the suspected fault cause; Step 211: Execute the fault analysis solution to obtain credibility parameters; Step 212: When the credibility parameter is higher than a preset credibility threshold, output the suspected fault cause corresponding to the credibility parameter.

[0007] Optionally, also include: Step 22: defining the number of suspected failure probabilities as the number of suspected failures; Step 220: If the number of suspected faults is equal to 1, obtain a fault analysis solution from the fault analysis solution library based on the suspected fault cause; Step 221: If the number of suspected faults is greater than 1, all suspected fault causes are combined based on the number of suspected faults to obtain a suspected fault cause group; Step 2210: Inputting the historical signal strength information into a preset multi-source fault analysis model to obtain a joint fault probability; Step 2211: taking the joint failure probability with the highest value as the suspected joint failure probability; Step 2212: When the suspected fault cause corresponding to the suspected joint fault probability is the same as the suspected fault cause corresponding to the suspected fault cause group, obtaining a fault analysis solution from the fault analysis solution library based on the joint fault probability; Step 2213: Continue to execute step 211; Step 2214: When the suspected fault cause corresponding to the suspected joint fault probability is different from the suspected fault cause corresponding to the suspected fault probability, the fault cause corresponding to the fault probability with the largest value is output as the suspected fault cause.

[0008] Optionally, the method further includes obtaining a fault analysis solution from the fault analysis solution library when the fault probability does not exceed the fault probability threshold, the method comprising: Step 222: Obtain corresponding standard fault phenomena from a preset fault analysis database according to the fault cause; Step 223: performing correlation analysis on all the standard fault phenomena to obtain similarity parameters between all the fault causes, wherein the similarity parameters represent the similarity of the standard fault phenomena between different fault causes; Step 224: When the similarity parameter is greater than a preset similarity threshold, the corresponding fault causes are integrated into a group of fault causes with the same phenomenon; Step 225: Accumulate the fault probabilities corresponding to all the fault causes in the same fault cause group to obtain an accumulated fault probability; Step 2250: If the accumulated fault probability is greater than a preset accumulated probability threshold, all the fault causes in the same fault cause group are regarded as suspected fault causes and step 210 is continued. Step 2251: If the accumulated failure probability does not exist and is greater than the accumulated probability threshold, execute step 20.

[0009] Optionally, the method further includes obtaining a fault analysis solution from the fault analysis solution library when the cumulative fault probability does not exist and is greater than the cumulative probability threshold, the method comprising: Step 22510: Analyze the actual characteristic parameter change trends corresponding to all preset characteristic information types based on the historical data information; Step 22511: Obtaining the change trends of all standard characteristic parameters corresponding to the fault cause based on the standard fault phenomenon; Step 22512: traverse all the fault causes to obtain a multi-source fault cause group, wherein the variation trends of all the standard characteristic parameters in the multi-source fault cause group are the same as the variation trends of the actual characteristic parameters after superposition; Step 22513: Obtain a fault analysis solution from the fault analysis solution library based on all the fault causes in the multi-source fault cause group and output it.

[0010] Optionally, a fault analysis method is further included when the multi-source fault cause group does not exist, the method comprising: Step 225120: Obtain the generation time, generation location, and implementation environment parameters of the fault signal; Step 225121: Searching for historical fault events, historical fault causes, and fault occurrence frequencies under the conditions of the occurrence time and the occurrence location from a preset historical fault cause table; Step 225122: Retrieve historical environmental parameters corresponding to the historical fault event; Step 225123: performing similarity analysis on the implementation environment parameters and the historical environment parameters to obtain environment similarity parameters; Step 225124: When the environment similarity parameter is greater than a preset environment similarity threshold, defining the corresponding historical fault event as a suspected fault event; Step 225125: Output the historical fault cause corresponding to the suspected fault event with the highest fault occurrence frequency as the suspected fault cause.

[0011] Optionally, a method for obtaining and outputting the credibility parameter is further included, the method comprising: Step 2110: Obtain transmission data information of the route to which the fault signal belongs; Step 21100: When the transmission data information is not obtained, the fault signal is defined as a data acquisition abnormality signal; Step 211001: searching for all historical characteristic parameter change trends corresponding to the data acquisition abnormal signal based on a preset data cause table; Step 211002: performing similarity analysis on all the historical characteristic parameter change trends and the actual characteristic parameter change trends to obtain trend similarity parameters; Step 211003: defining the trend similarity parameter with the largest value as the most similar trend parameter; Step 211004: Obtain an anomaly correction parameter from a preset acquisition anomaly correction parameter table according to the most similar trend parameter; Step 211005: Obtain an abnormality correction credibility parameter according to the credibility parameter and the abnormality correction parameter, and output it as the credibility parameter; Step 21101: When the transmission data information is obtained, the fault signal is defined as a data integrity abnormality signal; Step 211010: Obtain complete data information from the preset backup line; Step 211011: Obtaining a data loss reason from a preset data loss principle table based on the complete data information and the transmitted data information; Step 211012: Obtain integrity correction parameters from a preset data integrity parameter table based on the data loss cause and the suspected fault cause; Step 211013: Obtain an integrity correction credibility parameter based on the credibility parameter and the integrity correction parameter, and output it as the credibility parameter.

[0012] Optionally, the method for obtaining and outputting the credibility parameter further includes: Step 21102: when the data integrity abnormality signal and the data acquisition abnormality signal are received alternately during a preset historical period, defining the fault signal as a mixed abnormality signal; Step 211020: Analyze the proportion of the data integrity abnormal signal and the data acquisition abnormal signal in the mixed abnormal signal in the historical time, and define it as the mixed abnormality proportion; Step 211021: Simulate fault results based on the suspected fault causes corresponding to all the fault analysis solutions to obtain a simulated abnormal signal; Step 211022: Analyze the proportion of the data integrity abnormal signal and the data acquisition abnormal signal in the simulated abnormal signal in the historical time, and define it as the simulated abnormal proportion; Step 211023: obtaining a mixing correction parameter from a preset mixing correction database according to the ratio of the mixing anomaly proportion to the simulated anomaly proportion; Step 211024: Obtain a mixed correction credibility parameter based on the credibility parameter and the mixed correction parameter, and continue to execute step 212 using the mixed correction credibility parameter as the credibility parameter.

[0013] Optionally, the diagnostic method when the credibility parameter is lower than the credibility threshold comprises: Step 2120: Sort the suspected fault causes based on the fault probabilities to obtain a suspected fault cause ranking table; Step 2121: traverse the suspected fault cause ranking table, and define the suspected fault cause next to the suspected fault cause as the next suspected fault cause; Step 21210: When the credibility parameter of the fault analysis solution corresponding to the suspected fault cause is lower than the credibility threshold, the second suspected fault cause is taken as the suspected fault cause and the process continues with step 2121; Step 21211: When the credibility parameter of the fault analysis solution corresponding to the suspected fault cause is higher than the credibility threshold, output the suspected fault cause corresponding to the credibility parameter.

[0014] Optionally, the diagnosis method when the credibility parameter of the fault analysis solution corresponding to the absence of the suspected fault cause is higher than the credibility threshold includes: Step 212100: Find all fault device points preset in the fault analysis plan; Step 212101: Obtain a fault analysis solution group based on integrating the fault analysis solutions corresponding to the faulty device points; Step 212102: Accumulate all the credibility parameters of the fault analysis solution group to obtain the device point credibility; Step 212103: When the device point credibility exceeds a preset device credibility threshold, obtaining a possible fault cause of the faulty device point corresponding to the device point credibility; Step 212104: Output the possible fault cause as the suspected fault cause.

[0015] Optionally, a method for maintaining the fault analysis solution library includes: Step 3: After receiving the preset fault repair signal, obtain the fault cause and fault phenomenon corresponding to the fault signal; Step 30: Input the fault signal into the fault analysis solution library to obtain a corresponding fault ID; Step 31: Input the fault cause, the fault phenomenon, and the historical data information corresponding to the fault phenomenon into the fault analysis solution library according to the fault ID.

[0016] In summary, this application includes at least one of the following beneficial technical effects: The full-chain automation design, including automatic collection of historical data after a fault signal is triggered, calling a pre-trained model to calculate the fault probability, dynamically adjusting the sorting of suspected causes, and outputting results based on credibility parameters, significantly reduces the need for manual intervention.

[0017] The method innovatively integrates multi-level analysis logic such as independent analysis of single faults, calculation of joint probability of multiple faults, clustering of fault groups with the same phenomenon, and matching of multi-source feature trends. It fully covers complex scenarios such as single faults, concurrent multiple faults, and hidden faults, breaking through the limitations of traditional single-dimensional diagnosis.

[0018] The credibility of the results is ensured through the dual mechanism of forward analysis and reverse verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of a HPLC fault diagnosis method in an embodiment of the present application.

[0020] Figure 2 This is a flowchart of a diagnostic method when a credibility parameter is lower than a credibility threshold in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0022] The present application embodiment discloses a method for HPLC fault diagnosis. Figure 1 , a HPLC fault diagnosis method comprises: Step 1: In response to a fault signal, obtain historical data information on the line corresponding to the fault signal.

[0023] The fault signal is an abnormal event identifier used to trigger the fault diagnosis process in the HPLC system, which comes from the active reporting of the HPLC terminal.

[0024] Historical data information refers to the relevant data on the fault signal route within a preset time period before the fault occurs, which includes key parameters such as bit error rate (BER), received signal strength (RSSI) and signal-to-noise ratio (SNR).

[0025] By extracting the communication log from the local storage of the HPLC terminal device, the historical data information of the preset time window is traced back according to the timestamp of the fault signal. For example, if the fault occurs at 10:00:00, the data from 09:50:00 to 10:05:00 is extracted.

[0026] Step 2: Input the historical data information into the preset fault analysis model to obtain the failure probabilities corresponding to all preset fault causes.

[0027] The fault analysis model is derived from historical data and is used to determine the cause of the current fault and the corresponding failure probability. This model is constructed using machine learning methods: Three core parameters, namely bit error rate, received signal strength, and signal-to-noise ratio, are obtained from previous faults and combined with the fault labels to form a sample set. The dataset is then divided into a 70% training set, a 15% test set, and a 15% validation set, and the data is normalized to eliminate dimensional differences. Algorithms suitable for small sample classification, such as decision trees and random forests, are then selected to learn the mapping between different fault causes and parameter characteristics based on the training set. The model hyperparameters are then optimized using the validation set. Finally, the test set is used to evaluate the model's generalization ability, and the optimal model with a balanced precision and recall is selected as the final analysis model.

[0028] Fault causes are pre-defined, specific factors that may cause HPLC communication anomalies. These factors are typically derived from expert experience, historical failure cases, and theoretical analysis methods. These factors include environmental interference, physical link problems, terminal device failures, and multipath effects in complex networks.

[0029] The failure probability is a quantitative value of the confidence level of each preset failure cause in the fault analysis model, indicating the possibility that the model believes that the cause has caused the current failure.

[0030] Step 20: Outputting a preset false fault signal when all the fault probabilities do not exceed a preset fault probability threshold.

[0031] The failure probability threshold is a preset numerical threshold for failure probability, used to determine the credibility of the fault cause output by the fault analysis model. This threshold is determined through extensive experimentation by staff. By evaluating the degree of consistency between the failure probability and actual failure occurrence, the lowest value that demonstrates consistency between the failure probability and actual failure occurrence is used as the failure probability threshold.

[0032] The false fault signal is a signal automatically generated by the system when the failure probability of all preset fault causes in the fault diagnosis system does not exceed the preset threshold value, which is used to indicate that the current fault analysis process has failed to locate a credible fault cause.

[0033] If all fault probabilities do not exceed the preset fault probability threshold, this indicates that the existing fault analysis model cannot find a typical pattern that matches the current fault signal in the preset fault cause library. This may be due to insufficient correlation between historical data characteristics and the fault cause, resulting in an inaccurate analysis, or it may be a false alarm. Therefore, the system determines that the current diagnostic process cannot reach a reliable conclusion and outputs an erroneous fault signal to trigger manual intervention or expand the scope of investigation.

[0034] Step 21: When the failure probability exceeds the failure probability threshold, the corresponding failure probability is defined as a suspected failure probability, and the failure cause corresponding to the suspected failure probability is defined as a suspected failure cause.

[0035] The suspected fault probability refers to the probability value corresponding to the fault cause output by the fault analysis model that exceeds the preset fault probability threshold. The suspected fault cause refers to the fault cause output by the fault analysis model that exceeds the preset fault probability threshold.

[0036] If the fault probability exceeds the fault probability threshold, it means that the model has identified a potential root cause that has a high degree of match with the current fault signal based on historical data characteristics, such as the changing trends of parameters such as bit error rate (BER), received signal strength (RSSI), and signal-to-noise ratio (SNR). The confidence level has reached the preset standard and it will enter the subsequent analysis process as a credible clue.

[0037] Step 210: Obtain a fault analysis solution from a preset fault analysis solution library based on the suspected fault cause.

[0038] The Fault Analysis Solution Library is a collection of pre-configured verification solutions for various suspected fault causes. It stores detection methods, data collection methods, and judgment rules that correspond to each fault cause. The Fault Analysis Solution Library is obtained by analyzing a large number of historical actual faults. The staff correlates the fault causes with changes in environmental and device parameters. Based on the operating scenarios, they then develop detection solutions that capture these environmental and device parameters. Ultimately, they establish a correlation between the fault cause, environmental and device parameters, and detection solutions, and store this correlation in the Fault Analysis Solution Library.

[0039] Fault analysis solutions are verification solutions for specific suspected fault causes retrieved from a library. By entering the suspected fault cause code or keyword into the Fault Analysis Solution Library, the system can quickly match and retrieve the corresponding detection solution.

[0040] Step 211: Execute the fault analysis solution to obtain the reliability parameters.

[0041] The credibility parameter is a quantitative indicator generated after executing the fault analysis plan, which is used to measure the match between the suspected fault cause and the actual situation. According to the requirements of the fault analysis plan, the system collects or retrieves relevant environmental parameters such as temperature, humidity, electromagnetic interference, and equipment parameters such as voltage, current, and signal strength in real time in the equipment operation scenario. The actual collected parameters are then compared with the expected normal value range or typical fault characteristic values ​​defined in the plan to quantify the degree of match between the suspected cause and the actual situation, and finally generate a credibility parameter to measure the possibility that the fault cause is true. For example, when the suspected fault cause is determined to be the temperature exceeding the sensor calibration threshold, the system calculates the cosine similarity between the temperature parameters under the current conditions and the preset typical temperature parameters to generate a credibility parameter to measure the credibility of the suspected fault cause.

[0042] Step 212: When the credibility parameter is higher than a preset credibility threshold, output the suspected fault cause corresponding to the credibility parameter.

[0043] The credibility threshold is a preset numerical threshold used to determine whether the credibility parameters obtained after executing the fault analysis plan meet the credibility standard. The credibility threshold is set directly by the staff.

[0044] When the credibility parameter exceeds the credibility threshold, the fault cause is verified as credible. The real-time data or evidence obtained through the fault analysis solution highly matches the typical characteristics of the suspected fault cause, significantly reducing the probability of the model misjudging. Therefore, the system considers the authenticity of the current fault cause to have been verified across multiple dimensions and can be output as the fault cause.

[0045] Also includes: Step 22: Define the number of suspected failure probabilities as the number of suspected failures.

[0046] The number of suspected faults refers to the total number of fault causes screened out by the fault analysis model and whose failure probability exceeds the preset threshold.

[0047] Step 220: If the number of suspected faults is equal to 1, obtain a fault analysis solution from a fault analysis solution library based on the suspected fault cause.

[0048] When the number of suspected faults is equal to 1, it indicates that the model has converged to a unique fault cause with high confidence. At this time, the corresponding solution can be directly obtained from the fault analysis solution library through the table lookup method.

[0049] Step 221: If the number of suspected faults is greater than 1, all suspected fault causes are combined based on the number of suspected faults to obtain a suspected fault cause group.

[0050] When the number of suspected faults is greater than 1, it indicates that the system may have multiple independent fault sources, such as a power failure and hardware damage occurring simultaneously, or a single fault causing multi-dimensional anomalies, such as a loose antenna causing signal attenuation and increased bit error rate.

[0051] The suspected fault cause group refers to a possible fault combination set formed by combining all suspected fault causes when the number of suspected faults is greater than 1.

[0052] Step 2210: Input the historical signal strength information into a preset multi-source fault analysis model to obtain a joint fault probability.

[0053] The joint failure probability refers to the overall failure probability value obtained by comprehensive calculation when multiple suspected failure causes occur simultaneously or act in correlation with each other.

[0054] The multi-source fault analysis model is constructed using machine learning methods. Specifically, the model is constructed by analyzing the changing trends of three core parameters, namely bit error rate (BER), received signal strength (RSS), and signal-to-noise ratio (SNR), during past faults, and combining these with fault labels to form a sample set. During this sample set construction, the nonlinear coupling characteristics of multiple fault parameters are taken into account. When selecting fault causes, typical cases where multiple faults occur in combination are prioritized, such as equipment aging combined with electromagnetic interference, or line loss with signal attenuation. Deep learning algorithms are then used to discover implicit association rules between different fault parameters.

[0055] Step 2211: The joint failure probability with the highest value is taken as the suspected joint failure probability.

[0056] The suspected joint fault probability is the highest joint probability value among all possible combinations of fault causes calculated by the multi-source fault analysis model in a concurrent multi-fault analysis scenario. This parameter represents the combined probability of the most likely combination of faults. It is the optimal solution selected by the system from multiple potential fault relationships and is used to guide subsequent fault verification and repair decisions. The model construction method is the same as in step 2 and is not detailed here.

[0057] Step 2212: When the suspected fault cause corresponding to the suspected joint fault probability is the same as the suspected fault cause corresponding to the suspected fault cause group, a fault analysis solution is obtained from the fault analysis solution library based on the joint fault probability.

[0058] When the suspected fault cause corresponding to the suspected joint fault probability is the same as the suspected fault cause corresponding to the suspected fault probability group, it indicates that the current fault signal is likely caused by multiple fault causes. Since both the fault analysis model and the multi-source fault analysis model yield the same result, the credibility is high, and the suspected fault causes corresponding to the joint fault probability can be analyzed as possible causes. The suspected fault causes corresponding to the suspected joint fault probability and the suspected fault causes corresponding to the suspected fault cause group are both sets of suspected roadblock causes composed of multiple suspected fault causes. When making comparisons, the number of causes must be consistent and the specific causes must match completely.

[0059] Step 2213: Continue to execute step 211.

[0060] Step 2214: When the suspected fault cause corresponding to the suspected joint fault probability is different from the suspected fault cause corresponding to the suspected fault probability, the fault cause corresponding to the fault probability with the largest value is output as the suspected fault cause.

[0061] When the suspected fault cause corresponding to the suspected joint fault probability is inconsistent with the suspected fault cause corresponding to the suspected fault probability, it means that the model cannot clearly determine whether the fault is caused by a single cause or multiple causes. Therefore, the system first selects the fault cause corresponding to the single fault probability with the largest value as the decision logic for simplifying complex scenarios of the suspected fault cause, and prioritizes analyzing the cause with the highest probability to ensure timely handling of the fault.

[0062] Also included is a method for obtaining a fault analysis solution from a fault analysis solution library when the probability of no fault exceeding a fault probability threshold exists, the method comprising: Step 222: Obtain corresponding standard fault phenomena from a preset fault analysis database according to the fault cause.

[0063] The fault analysis database, obtained through extensive analysis, contains fault causes and their corresponding classic fault phenomena. This database collects fault causes from past operations and compiles the corresponding fault phenomena for each cause. For example, by analyzing numerous external electromagnetic interference (EMI) faults, the staff concluded that under these conditions, the bit error rate (BER) would increase significantly, the RSSI (Reverse Signal-to-Noise) would fluctuate and decrease, the signal-to-noise ratio (SNR) would drop significantly, and the noise level would increase. These key parameter changes were used as the standard fault phenomena corresponding to this cause. A corresponding relationship between the fault cause and the standard fault phenomena was established and stored in the fault analysis database.

[0064] Standard fault phenomena are observable characteristics corresponding to typical faults verified in the historical operation of a device or system, such as temperature anomalies and response delays. They serve as a reference for current fault diagnosis. By entering the cause of a fault into the fault analysis database, the corresponding standard fault phenomenon can be obtained.

[0065] Step 223: performing correlation analysis on all standard fault phenomena to obtain similarity parameters between all fault causes. The similarity parameters represent the similarity of standard fault phenomena between different fault causes.

[0066] Correlation analysis uses statistical or algorithmic methods to calculate the similarity between standard fault phenomena corresponding to fault causes. By obtaining the parameter variation characteristics of all preset key parameters corresponding to the standard fault phenomena, key parameters with identical variations between the two standard fault phenomena are recorded as similar parameters. The similarity parameter can be calculated by dividing the number of similar parameters by the number of key environmental parameters and equipment parameters. For example, the key environmental and equipment parameters are: bit error rate, received signal strength, and signal-to-noise ratio. Fault A is power line interference, Fault B is impedance mismatch, and Fault C is external electromagnetic interference. Fault A exhibits an increase in BER, a decrease in RSSI, and a decrease in SNR. Fault B exhibits no effect on BER, a decrease in RSSI, and a decrease in SNR. Fault C exhibits an increase in BER, a decrease in RSSI, and a decrease in SNR. If Faults A and B share the same parameters of RSSI and SNR, and the similarity parameter is 2 / 3, the phenomena are relatively similar, with both affecting RSSI and SNR. If Faults A and C share the same parameters of BER, RSSI, and SNR, and the similarity parameter is 1, the phenomena are identical. This is just an example; the specific key environmental and equipment parameters should be determined based on on-site environmental conditions and equipment usage.

[0067] The similarity parameter is a parameter that quantitatively describes the degree of similarity between different fault causes at the level of standard fault phenomena. The larger the similarity parameter, the more similar the fault phenomena caused by the faults are.

[0068] Step 224: When the similarity parameter is greater than a preset similarity threshold, the corresponding fault causes are integrated into a group of fault causes with the same phenomenon.

[0069] The similarity threshold is determined by analyzing a large number of historical fault phenomena. The threshold is the lowest value that indicates a high degree of similarity between the fault phenomena caused by two fault causes. When the similarity parameter exceeds the preset threshold, it indicates that the standard fault phenomena caused by these fault causes, such as BER, RSSI, and SNR, have a high degree of overlap in key characteristics, possibly due to similar physical mechanisms or external interference. If each fault cause is analyzed separately, the probability of each cause being "equalized" due to the similarity of the phenomena.

[0070] A common fault cause group is formed by combining multiple fault causes whose similarity parameters exceed a preset similarity threshold. When multiple fault causes lead to highly similar standard fault phenomena, these causes are difficult to distinguish directly at the phenomenological level, but their impact on the system is consistent, so they need to be combined into a group for joint analysis.

[0071] Step 225: Accumulate the failure probabilities corresponding to all the failure causes in the same failure phenomenon group to obtain the accumulated failure probability.

[0072] The cumulative failure probability is the total probability value obtained by summing the failure probabilities of all failure causes within a group of failure causes for the same phenomenon. By accumulating the independent probabilities of each failure cause within a group, we can avoid underestimating the probability of a single cause due to similar phenomena, thereby more reliably judging the overall credibility of the phenomenon.

[0073] Step 2250: If the accumulated fault probability is greater than the preset accumulated probability threshold, all fault causes in the same fault phenomenon group are considered as suspected fault causes and step 210 is continued.

[0074] The cumulative probability threshold is obtained by staff through analyzing a large number of actual historical faults. It is the minimum parameter that can prove that the fault phenomenon is actually credible. The staff defines this parameter as the cumulative probability threshold.

[0075] A cumulative failure probability greater than the preset cumulative probability threshold indicates that the standard failure phenomena caused by all fault causes within the same fault phenomenon group are not accidental. This phenomenon occurs because the standard failure phenomena caused by multiple fault causes are highly similar, which affects the judgment of the fault cause. However, a cumulative failure probability greater than the threshold indicates that the fault phenomenon is confirmed to be credible. Therefore, to identify the fault cause, all fault causes must be treated as suspected fault causes for the subsequent verification steps. For example, strong external electromagnetic interference and power line aging can both cause RSSI drop and BER increase. The calculated failure probabilities for each are 30% and 35%, respectively. However, the similarity of the two phenomena can actually lead to a misjudgment. The actual probability of the "RSSI drop and BER increase" fault phenomenon reaches 65%, so careful analysis of this fault phenomenon is necessary.

[0076] Step 2251: If there is no cumulative failure probability greater than the cumulative probability threshold, execute step 20.

[0077] If the cumulative fault probability corresponding to the non-existent fault phenomenon group exceeds the cumulative probability threshold, it indicates that the correlation between the currently observed fault phenomenon and the known fault causes is low. This indicates that the probabilities of all fault causes are low, not due to similar fault phenomena causing the probability of a single cause to be dispersed. In this case, the system determines that in-depth cause analysis of the existing fault phenomenon group is unnecessary and directly triggers step 20.

[0078] A method for obtaining a fault analysis solution from a fault analysis solution library when there is no cumulative fault probability greater than a cumulative probability threshold, the method comprising: Step 22510: Analyze the actual characteristic parameter change trends corresponding to all preset characteristic information types based on historical data information.

[0079] Feature information types refer to observable parameter categories that describe fault phenomena, typically key indicators of device operating status, such as bit error rate, received signal strength, signal-to-noise ratio, and delay jitter.

[0080] The actual characteristic parameter change trend refers to the direction of parameter change corresponding to the characteristic information type in an actual fault scenario. For example, BER increases, RSSI remains unchanged, and SNR decreases.

[0081] The system extracts time series data of target feature parameters from historical data, arranges them chronologically, removes outliers, and iterates through the data points, counting increases and decreases between adjacent points. If the current value exceeds the previous value, it's recorded as an "increase," otherwise, it's recorded as a "decrease," and if they're equal, it's recorded as "no change." If the percentage of "increases" exceeds a set threshold, such as 70%, and there's no significant reverse fluctuation, it's considered monotonically increasing. Similarly, if the percentage of "decreases" exceeds a threshold, it's considered monotonically decreasing. If the percentage of "no change" exceeds a threshold, such as 80%, it's considered stable. If increases and decreases alternate frequently with no dominant direction, such as if the percentage of increases and decreases approaches 50%, it's considered fluctuating. The percentage threshold can vary depending on the actual application environment and is provided here for illustrative purposes only.

[0082] Step 22511: Obtain the change trends of all standard characteristic parameters corresponding to the fault cause based on the standard fault phenomenon.

[0083] Standard characteristic parameter change trends refer to the qualitative or quantitative change patterns of characteristic parameters corresponding to the fault cause, as summarized by analyzing the correlation between typical fault phenomena and the fault cause. For example, when a certain type of fault occurs, specific parameters may show a trend of monotonically increasing, decreasing, fluctuating, or showing no significant change. For example, because power line carrier signals rely on voltage amplitude for transmission, a voltage drop will cause a significant drop in signal strength, manifesting as a sudden drop in RSSI values.

[0084] Step 22512: traverse all fault causes to obtain a multi-source fault cause group, and the change trends of all standard characteristic parameters in the multi-source fault cause group are the same as the change trends of the actual characteristic parameters after superposition.

[0085] A multi-source fault cause group is a collection of two or more fault causes. The standard characteristic parameter trends corresponding to all of these fault causes, such as increases, decreases, or fluctuations, are superimposed or interact with each other, resulting in an overall effect that fully matches the observed characteristic parameter trends. For example, in an HPLC communication system, if RSSI decreases and BER increases, the multi-source fault group may be composed of two fault causes: unstable power supply voltage and line interference. The former causes RSSI decreases, while the latter causes BER increase. The combined trend of these two factors fully matches the actual phenomenon.

[0086] Step 22513: Obtain a fault analysis solution from the fault analysis solution library based on all fault causes in the multi-source fault cause group and output it.

[0087] By inputting all the fault causes in the multi-source fault cause group into the fault analysis solution library, the fault analysis solution corresponding to each fault cause can be obtained, and then the fault analysis solution can be used to determine whether the analyzed fault cause is credible.

[0088] Also included is a fault analysis method when there is no multi-source fault cause group, the method comprising: Step 225120: Retrieve the generation time, generation location and implementation environment parameters of the fault signal.

[0089] The occurrence time refers to the specific time node when the fault occurs. It only records the specific time when the fault occurs. For example, if the fault occurs at 12:35, the date is not recorded.

[0090] The location of a fault is the physical or logical location where the fault occurred. This can be the specific area or network node where the equipment is deployed. For example, the location of a distribution automation terminal fault is "Distribution Cabinet 3, Substation XX, City XX, Province XX," or the location of an HPLC communication fault is "Phase A power line, GPS coordinates (longitude 120.12°, latitude 30.56°)."

[0091] Implementation environmental parameters refer to the real-time or historical conditions of the environment in which the equipment is located when the fault occurs, including temperature, humidity, voltage, current, electromagnetic interference, weather conditions and other parameters that directly affect the operation of the equipment.

[0092] Step 225121: Search the preset historical fault cause table for historical fault events, historical fault causes, and fault occurrence frequencies under the conditions of occurrence time and occurrence location.

[0093] The historical fault cause table is a structured data table constructed by staff after analyzing a large number of historical fault cases. It contains key information about historical fault events, such as the time, location, and environmental parameters, as well as the corresponding fault cause type and frequency. Staff search for historical fault events within the system, link the key information corresponding to the fault with the fault event, and accumulate the number of faults in chronological order. The fault frequency is calculated by dividing the number of fault causes by the total number of faults.

[0094] Simply input the occurrence time and location into the historical fault cause table, and the system will obtain the historical fault time that occurred at that time and location under historical conditions, and find the historical fault cause and fault frequency corresponding to the event based on the historical fault cause table.

[0095] Step 225122: Retrieve historical environmental parameters corresponding to historical fault events.

[0096] Historical environmental parameters refer to quantified data about the external environment or equipment operating status associated with historical fault events, describing the objective conditions at the time of the fault. These parameters include physical environment parameters such as temperature, humidity, and air pressure, and electrical environment parameters such as voltage fluctuations, current anomalies, and electromagnetic interference intensity.

[0097] By inputting the ID of a historical fault event from the system, the historical environmental parameters corresponding to the fault time can be retrieved from the system.

[0098] Step 225123: Perform similarity analysis on the implementation environment parameters and the historical environment parameters to obtain environment similarity parameters.

[0099] The environmental similarity parameter is a numerical indicator that reflects the similarity between the current fault environment parameters (such as temperature, humidity, voltage, and electromagnetic interference) and the historical fault environment parameters. For example, the current environmental parameters are: temperature 45°C, humidity 30%, and voltage fluctuation ±8%. The historical environmental parameters are: temperature 43°C, humidity 28%, and voltage fluctuation ±10%. The assumed parameter range is: temperature 0-100°C, humidity 0-100%, and voltage fluctuation 0-20%.

[0100] After normalization, the current temperature is normalized to 45 / 100 = 0.45, and the historical temperature is normalized to 43 / 100 = 0.43. The current humidity is normalized to 30 / 100 = 0.3, and the historical humidity is normalized to 28 / 100 = 0.28. The current voltage is normalized to 8 / 20 = 0.4, and the historical voltage is normalized to 10 / 20 = 0.5.

[0101] Substitute the above values ​​into the Euclidean distance formula to obtain the Euclidean distance, and then substitute the distance into the similarity calculation formula: Substituting the numerical value, the similarity obtained is 0.94, which means that the similarity parameter between the current environment and the historical environment is 94%.

[0102] Step 225124: When the environment similarity parameter is greater than a preset environment similarity threshold, the corresponding historical fault event is defined as a suspected fault event.

[0103] A suspected fault event refers to a fault event that may have the same cause as a historical fault event.

[0104] The environmental similarity threshold is determined by personnel through analysis of numerous past failure events. It represents the minimum value that demonstrates the similarity between current environmental parameters and historical failure parameters. When the environmental similarity parameter exceeds the preset threshold, it indicates a close match between the current environment and the historical failure environment. This indicates that the current device operating environment, such as temperature, humidity, and voltage fluctuations, is sufficiently similar to the environmental conditions during a historical failure to trigger the same or similar failure mechanism. Therefore, the corresponding historical failure event can be treated as a suspected failure event.

[0105] Step 225125: Output the historical fault cause corresponding to the suspected fault event with the highest fault occurrence frequency as the suspected fault cause.

[0106] By counting the frequency of suspected fault events in historical data, the historical fault causes corresponding to the events with the highest frequency are selected as the output results. Its core logic is to use the statistical laws of historical data to prioritize the fault causes with the highest probability of occurrence.

[0107] The method further includes obtaining and outputting a credibility parameter, the method comprising: Step 2110: Obtain transmission data information of the route to which the fault signal belongs.

[0108] Transmission data refers to the raw data on the communication line or data transmission path associated with the fault signal. This data can be retrieved from archived transmission data on that route before the fault occurred by querying a historical database or storage system.

[0109] Step 21100: When the transmission data information is not obtained, the fault signal is defined as a data acquisition abnormality signal.

[0110] A data collection anomaly signal is an abnormal event flagged by the system during the fault diagnosis process due to a failure to obtain transmitted data information. This signal indicates an issue with the collection process or data source, not the data content itself.

[0111] If no transmission data information is obtained, it indicates that there is a defect in the data collection process, which may be due to equipment failure or human error, resulting in the system being unable to extract the required transmission data from the specified route.

[0112] Step 211001: Search for all historical characteristic parameter change trends corresponding to data acquisition abnormal signals based on a preset data cause table.

[0113] The data cause table is a pre-set structured database or knowledge base that stores the mapping between data acquisition anomalies and corresponding historical characteristic parameter change trends. This table is obtained by staff searching through a large amount of historical failure data. Staff correlate the changing trends of parameters such as bit error rate, received signal strength, signal-to-noise ratio, and delay jitter during the transmission data acquisition failure process, store them in the data cause table, and label them as data acquisition anomalies.

[0114] By inputting the label of the data acquisition abnormal signal into the data cause table, the change trend of each historical characteristic parameter at the time of all failures caused by the failure to obtain transmission data information can be obtained.

[0115] Step 211002: Perform similarity analysis on all historical characteristic parameter change trends and actual characteristic parameter change trends to obtain trend similarity parameters.

[0116] The trend similarity parameter is a quantitative indicator used to measure the degree of match between historical and actual feature parameter change trends. This parameter outputs a normalized similarity value by comprehensively analyzing the changing patterns of multiple feature parameters (such as signal strength, bit error rate, and latency), combined with characteristics such as the time series shape, slope, and periodicity.

[0117] Similarity analysis is a method for evaluating the consistency or difference between two or more objects by quantitatively comparing their matching. By comparing the consistency of historical feature parameter change trends with actual feature parameter change trends and calculating the proportion of parameters with the same trend, the degree of similarity between the two can be quantitatively assessed. The specific calculation method can be found in step 223. Other similarity calculation methods, such as cosine similarity, are also available and will not be detailed here.

[0118] Step 211003: Define the trend similarity parameter with the largest value as the most similar trend parameter.

[0119] The most similar trend parameter refers to the characteristic parameter with the largest similarity value in the similarity analysis between the change trends of all historical characteristic parameters and the change trends of the actual characteristic parameters.

[0120] Step 211004: Obtain anomaly correction parameters from a preset acquisition anomaly correction parameter table based on the trend most similar parameters.

[0121] The anomaly correction parameter table is a preconfigured mapping table that stores different trend-similar parameters and their corresponding correction parameters. By analyzing a large number of historical faults, staff statistically analyze the correspondence between the most similar trend parameters and anomaly correction parameters. The larger the statistically most similar trend parameter, the larger the correction parameter, indicating a more reliable parameter. By entering the most similar trend parameter into the anomaly correction parameter table, the corresponding anomaly correction parameter can be found.

[0122] Step 211005: Obtain an abnormality correction credibility parameter based on the credibility parameter and the abnormality correction parameter, and output it as a credibility parameter.

[0123] The anomaly correction credibility parameter is a dynamic value calculated by combining the original credibility parameter and the acquisition anomaly correction parameter, which is used to quantify the reliability of the corrected conclusion.

[0124] The anomaly correction reliability parameter can be obtained by multiplying the original reliability parameter by the anomaly correction parameter.

[0125] Step 21101: When the transmission data information is obtained, the fault signal is defined as a data integrity abnormality signal.

[0126] A data integrity anomaly signal occurs when the system detects data loss, corruption, or incomplete structure during data transmission, causing the received data to fail to meet the preset integrity requirements. This signal is triggered by conditions such as packet loss, checksum errors, missing key fields, or abnormal values.

[0127] Step 211010: Obtain complete data information from the preset backup line.

[0128] A backup line is a redundant transmission channel or backup path pre-configured by the system. It is used to provide reliable and complete data supplementation when the data transmitted on the primary line has integrity issues. Complete data information is the data information that should have been received if the fault had not occurred.

[0129] When the data transmitted by the main line is detected as incomplete, the system automatically or manually switches to the backup line to obtain complete data information.

[0130] Step 211011: Obtain the data loss cause from a preset data loss principle table based on the complete data information and the transmitted data information.

[0131] The data loss principle table is a pre-set mapping table used to store association rules between different data loss patterns and their corresponding root causes. By analyzing a large number of historical failure events, staff have identified the corresponding relationships between different loss types and data loss causes in the case of data integrity anomalies. These relationships are stored in the data loss principle table.

[0132] By analyzing the missing types of data between the complete data information and the transmitted data information, such as field-level missing, packet-level missing, checksum-level anomaly, etc., and substituting the corresponding missing type into the data loss principle table, the possible data loss cause corresponding to the missing type can be obtained.

[0133] Step 211012: Obtain integrity correction parameters from a preset data integrity parameter table based on the cause of data loss and the suspected fault cause.

[0134] The data integrity parameter table is a pre-defined mapping table used to store integrity correction parameters corresponding to data loss causes and suspected fault causes. By analyzing a large number of fault events, staff compare the data loss causes obtained through table lookup with the suspected fault causes obtained through machine learning analysis. They then establish a correspondence between the similarity between the two causes and the integrity correction parameters, ensuring that the integrity correction parameters represent the similarity between the fault causes analyzed using the two methods.

[0135] For example, if the cause of data loss is identical to the suspected cause of a failure, the integrity correction parameter is set to 1.2. If the cause of data loss and the suspected cause of a failure are identical, the integrity correction parameter is set to 1.1. If the cause of data loss and the suspected cause of a failure are completely unrelated, the integrity correction parameter is set to 0.8. The above data is for demonstration purposes only. In actual use, the integrity correction parameter range and proposed parameters can be adjusted based on actual conditions.

[0136] Step 211013: Obtain the integrity correction credibility parameter based on the credibility parameter and the integrity correction parameter, and output it as the credibility parameter.

[0137] The integrity correction credibility parameter refers to a parameter value calculated based on the credibility parameter and the integrity correction parameter.

[0138] The integrity correction credibility can be obtained by multiplying the credibility parameter by the integrity correction parameter.

[0139] The method for obtaining and outputting the credibility parameter also includes: Step 21102: When a data integrity abnormality signal and a data acquisition abnormality signal are received alternately during a preset historical time, the fault signal is defined as a mixed abnormality signal.

[0140] A mixed abnormal signal refers to a composite abnormal state in which the system detects data integrity abnormal signals and data acquisition abnormal signals alternately within a preset historical time window, and the two have a temporal correlation.

[0141] When the system alternately detects data integrity anomalies and data collection anomalies within a historical time window, it indicates that the fault may be causing issues with both data integrity and collection capabilities. For example, when using HPLC technology to remotely collect smart meter data, some meters may experience missing voltage and current data during peak hours, which is a data integrity anomaly. Simultaneously, the concentrator may display "communication interruption," which is a data collection anomaly.

[0142] Step 211020: Analyze the proportion of data integrity abnormal signals and data acquisition abnormal signals in the mixed abnormal signals in the historical time, and define it as the mixed abnormality proportion.

[0143] The mixed anomaly ratio refers to the ratio of alternating occurrences of data integrity anomaly signals and data collection anomaly signals within a preset historical time window. It is used to quantify the degree of mixing between the two types of anomalies. The mixed anomaly ratio includes the historical proportion of data integrity anomaly signals and the historical proportion of data collection anomaly signals.

[0144] The proportion of mixed anomalies includes: {the proportion of data integrity anomaly signals, the proportion of data collection anomaly signals}. The proportion of data integrity anomaly signals can be obtained by dividing the number of data integrity anomaly signals by the number of samples in the historical time window, and the proportion of data collection anomaly signals can be obtained by dividing the number of data collection anomaly signals by the number of samples in the historical time window.

[0145] Step 211021: Simulate the fault results based on the suspected fault causes corresponding to all fault analysis solutions to obtain simulated abnormal signals.

[0146] The simulated anomaly signal is a theoretical anomaly pattern derived by reverse engineering the suspected fault cause. It is used to verify whether the actual anomaly is caused by that fault. A physical model of the anomaly is established based on the suspected fault cause. When the suspected fault is a transmission link damage, the bit error rate (BER) increases exponentially due to signal attenuation, the received signal strength (RSSI) decreases linearly with path loss, and the signal-to-noise ratio (SNR) decays logarithmically due to environmental noise. If the fault is an interference-related fault, the BER exhibits pulse-like mutations in the interference frequency band, the RSSI fluctuation variance increases, and the SNR plummets during the interference period. By matching parameter sensitivity coefficients with the fault mechanism library, the multi-dimensional parameter change trends are time-aligned and amplitude normalized, ultimately outputting a composite simulated anomaly signal that matches the fault characteristics.

[0147] Step 211022: Analyze the proportion of data integrity abnormal signals and data acquisition abnormal signals in the simulated abnormal signals in the historical time, and define it as the simulated abnormality proportion.

[0148] The simulated anomaly ratio refers to the relative proportion of data integrity anomalies and data collection anomalies in the abnormal signals generated by fault simulation within a preset historical time window.

[0149] For simulated anomaly signals, count the number of data integrity anomalies and data collection anomalies on a unified time scale. Compare this with the number of samples in the historical time window to obtain the simulated anomaly percentage. The calculation method for the simulated anomaly percentage is the same as in step 211020 and is not detailed here.

[0150] Step 211023: Obtain the mixed correction parameters from the preset mixed correction database according to the ratio of the mixed anomaly proportion to the simulated anomaly proportion.

[0151] The hybrid correction parameter is a parameter used to quantify the degree of match between actual anomaly patterns and simulated fault patterns. The hybrid correction database is a pre-set mapping table that stores hybrid correction parameters corresponding to different ratios of hybrid anomaly proportions to simulated anomaly proportions. By analyzing a large number of fault histories, staff correlated the proportions of hybrid anomaly proportions caused by different faults with the hybrid correction parameters. This allows the hybrid correction parameters to accurately reflect the similarity of anomaly proportions between different faults, and thus, the similarity between different faults.

[0152] Step 211024: Obtain a mixed correction credibility parameter based on the credibility parameter and the mixed correction parameter, and continue to execute step 212 using the mixed correction credibility parameter as the credibility parameter.

[0153] The hybrid correction credibility parameter refers to the parameter obtained by multiplying the credibility parameter by the hybrid correction parameter, and is used to quantify the matching degree of the suspected fault cause in the actual application.

[0154] The hybrid correction credibility parameter can be obtained by multiplying the credibility parameter by the hybrid correction parameter.

[0155] Reference Figure 2 , a diagnostic method when a credibility parameter is lower than a credibility threshold, the method comprising: Step 2120: Sort the suspected fault causes based on the fault probability to obtain a suspected fault cause sorting table.

[0156] The suspected fault cause ranking table prioritizes potential fault causes based on their probability. The system prioritizes the suspected fault causes by probability, sorting them from highest to lowest.

[0157] Step 2121: traverse the suspected fault cause ranking table, and define the suspected fault cause next to the suspected fault cause as the second suspected fault cause.

[0158] The next-highest suspected fault cause is the candidate fault cause that follows the current analysis target and is ranked next in the list of suspected fault causes sorted by fault probability. Its role is to serve as an alternative analysis target and continue in the fault diagnosis process if the credibility of the current fault cause does not meet the threshold.

[0159] Step 21210: When the credibility parameter of the fault analysis solution corresponding to the suspected fault cause is lower than the credibility threshold, the next suspected fault cause is taken as the suspected fault cause and step 2121 is continued.

[0160] When the credibility parameter of the fault analysis scheme corresponding to the suspected fault cause is lower than the threshold, it indicates that the current fault cause hypothesis lacks sufficient evidence to support it. Therefore, the next-highest priority with a slightly lower fault probability is selected as the suspected fault cause for verification.

[0161] Step 21211: When the credibility parameter of the fault analysis solution corresponding to the suspected fault cause is higher than the credibility threshold, the suspected fault cause corresponding to the credibility parameter is output.

[0162] When the credibility parameter of the fault analysis scheme corresponding to the suspected fault cause is higher than the credibility threshold, it indicates that the matching degree between the simulated abnormal signal of the fault cause and the actual observation data reaches the preset standard. The system terminates the traversal process and outputs the fault cause and its associated fault probability as the final diagnosis result, and ends the current fault analysis process at the same time.

[0163] A diagnostic method when a credibility parameter of a fault analysis solution corresponding to no suspected fault cause is higher than a credibility threshold, the method comprising: Step 212100: Find all fault device points preset in the fault analysis plan.

[0164] Predetermined fault points refer to specific hardware components, connection nodes, or functional modules in the HPLC system that are prone to communication failures due to physical structure, electrical characteristics, or environmental interactions. Examples include line connectors and connection points, power lines themselves, and filters.

[0165] Step 212101: Obtain a fault analysis solution group based on integrating the fault analysis solutions corresponding to the fault device points.

[0166] A fault analysis plan group is a collection of multiple fault analysis plans targeting the same fault location. This group is created by finding the potential fault locations within a fault analysis plan and then combining the corresponding fault analysis plans for the same potential fault location.

[0167] Step 212102: Accumulate all the credibility parameters of the fault analysis solution group to obtain the device point credibility.

[0168] The device point credibility is a comprehensive value obtained by accumulating the credibility parameters of all relevant fault analysis schemes for a certain fault device point, and is used to measure the possibility of failure of the device point.

[0169] Step 212103: When the device point credibility exceeds a preset device credibility threshold, a possible fault cause of the faulty device point corresponding to the device point credibility is obtained.

[0170] The device credibility threshold is a preset numerical threshold used to determine whether the credibility of a faulty device point meets the criteria for an alert or action. The device credibility threshold is determined through extensive testing and is the lowest threshold that indicates a high probability of a fault at the current faulty device point.

[0171] When the device point credibility exceeds a preset device credibility threshold, it is considered that the device point has a failure risk.

[0172] Possible fault causes refer to a set of specific fault hypotheses directly related to a faulty device point, determined by integrating the credibility parameters of all fault analysis schemes corresponding to the device point when the device point credibility of the device point exceeds a preset threshold.

[0173] Step 212104: Output the possible fault cause as a suspected fault cause.

[0174] A method for maintaining a fault analysis solution library, the method comprising: Step 3: After receiving the preset fault repair signal, obtain the fault cause and fault phenomenon corresponding to the fault signal.

[0175] The fault repair signal is a pre-set trigger mechanism in the system that indicates that a fault has been repaired or eliminated. This signal can be sent manually by maintenance personnel through the user interface after the fault is repaired, or it can be automatically generated by the system after sensors or logic determine that signal indicators have normalized.

[0176] The cause of a fault refers to the fundamental factor that causes a device or system abnormality, such as hardware damage (such as an aging power module), software vulnerabilities, signal interference, or excessive environmental parameters (such as excessive temperature).

[0177] Fault symptoms refer to the abnormal state of a device or system when a fault occurs, such as communication interruption, increased signal noise, device unresponsiveness, data packet loss, etc.

[0178] Step 30: Input the fault signal into the fault analysis solution library to obtain the corresponding fault ID.

[0179] The fault ID is a code used to uniquely identify a fault in the fault analysis solution library. After the staff enters key data such as the equipment status, alarm information, operation records, etc. at the time of each fault event into the fault analysis solution library, the system will automatically generate a unique fault ID based on the preset coding rules. This code serves as the digital identity of the fault in the system and is used to establish a fault file index, associate historical data, and conduct subsequent analysis and tracing.

[0180] Step 31: Input the fault cause, fault phenomenon, and historical data information corresponding to the fault phenomenon into the fault analysis solution library according to the fault ID.

Claims

1. A HPLC fault diagnosis method, characterized in that, include: Step 1: In response to a fault signal, obtaining historical data information on a line corresponding to the fault signal; Step 2: Inputting the historical data information into a preset fault analysis model to obtain the failure probabilities corresponding to all preset fault causes; Step 20: Outputting a preset error fault signal when all the fault probabilities do not exceed a preset fault probability threshold; Step 21: When the failure probability exceeds the failure probability threshold, the corresponding failure probability is defined as a suspected failure probability, and the failure cause corresponding to the suspected failure probability is defined as a suspected failure cause; Step 210: Obtaining a fault analysis solution from a preset fault analysis solution library based on the suspected fault cause; Step 211: Execute the fault analysis solution to obtain credibility parameters; Step 212: When the credibility parameter is higher than a preset credibility threshold, output the suspected fault cause corresponding to the credibility parameter.

2. A HPLC fault diagnosis method according to claim 1, characterized in that, Also includes: Step 22: defining the number of suspected failure probabilities as the number of suspected failures; Step 220: If the number of suspected faults is equal to 1, obtain a fault analysis solution from the fault analysis solution library based on the suspected fault cause; Step 221: If the number of suspected faults is greater than 1, all suspected fault causes are combined based on the number of suspected faults to obtain a suspected fault cause group; Step 2210: Inputting the historical signal strength information into a preset multi-source fault analysis model to obtain a joint fault probability; Step 2211: taking the joint failure probability with the highest value as the suspected joint failure probability; Step 2212: When the suspected fault cause corresponding to the suspected joint fault probability is the same as the suspected fault cause corresponding to the suspected fault cause group, obtaining a fault analysis solution from the fault analysis solution library based on the joint fault probability; Step 2213: Continue to execute step 211; Step 2214: When the suspected fault cause corresponding to the suspected joint fault probability is different from the suspected fault cause corresponding to the suspected fault probability, the fault cause corresponding to the fault probability with the largest value is output as the suspected fault cause.

3. A HPLC fault diagnosis method according to claim 2, characterized in that, The method further includes obtaining a fault analysis solution from the fault analysis solution library when the fault probability does not exceed the fault probability threshold, the method comprising: Step 222: Obtain corresponding standard fault phenomena from a preset fault analysis database according to the fault cause; Step 223: performing correlation analysis on all the standard fault phenomena to obtain similarity parameters between all the fault causes, wherein the similarity parameters represent the similarity of the standard fault phenomena between different fault causes; Step 224: When the similarity parameter is greater than a preset similarity threshold, the corresponding fault causes are integrated into a group of fault causes with the same phenomenon; Step 225: Accumulate the fault probabilities corresponding to all the fault causes in the same fault cause group to obtain an accumulated fault probability; Step 2250: If the accumulated fault probability is greater than a preset accumulated probability threshold, all the fault causes in the same fault cause group are regarded as suspected fault causes and step 210 is continued. Step 2251: If the accumulated failure probability does not exist and is greater than the accumulated probability threshold, execute step 20.

4. A HPLC fault diagnosis method according to claim 3, characterized in that, The method further includes obtaining a fault analysis solution from the fault analysis solution library when the accumulated fault probability does not exist and is greater than the accumulated probability threshold, the method comprising: Step 22510: Analyze the actual characteristic parameter change trends corresponding to all preset characteristic information types based on the historical data information; Step 22511: Obtaining the change trends of all standard characteristic parameters corresponding to the fault cause based on the standard fault phenomenon; Step 22512: traverse all the fault causes to obtain a multi-source fault cause group, wherein the variation trends of all the standard characteristic parameters in the multi-source fault cause group are the same as the variation trends of the actual characteristic parameters after superposition; Step 22513: Obtain a fault analysis solution from the fault analysis solution library based on all the fault causes in the multi-source fault cause group and output it.

5. A HPLC fault diagnosis method according to claim 4, characterized in that, Also included is a fault analysis method when the multi-source fault cause group does not exist, the method comprising: Step 225120: Obtain the generation time, generation location, and implementation environment parameters of the fault signal; Step 225121: Searching for historical fault events, historical fault causes, and fault occurrence frequencies under the conditions of the occurrence time and the occurrence location from a preset historical fault cause table; Step 225122: Retrieve historical environmental parameters corresponding to the historical fault event; Step 225123: performing similarity analysis on the implementation environment parameters and the historical environment parameters to obtain environment similarity parameters; Step 225124: When the environment similarity parameter is greater than a preset environment similarity threshold, defining the corresponding historical fault event as a suspected fault event; Step 225125: Output the historical fault cause corresponding to the suspected fault event with the highest fault occurrence frequency as the suspected fault cause.

6. A HPLC fault diagnosis method according to claim 4, characterized in that, Also included is a method for obtaining and outputting the credibility parameter, the method comprising: Step 2110: Obtain transmission data information of the route to which the fault signal belongs; Step 21100: When the transmission data information is not obtained, the fault signal is defined as a data acquisition abnormality signal; Step 211001: searching for all historical characteristic parameter change trends corresponding to the data acquisition abnormal signal based on a preset data cause table; Step 211002: performing similarity analysis on all the historical characteristic parameter change trends and the actual characteristic parameter change trends to obtain trend similarity parameters; Step 211003: defining the trend similarity parameter with the largest value as the most similar trend parameter; Step 211004: Obtain an anomaly correction parameter from a preset acquisition anomaly correction parameter table according to the trend most similar parameter; Step 211005: Obtain an abnormality correction credibility parameter according to the credibility parameter and the abnormality correction parameter, and output it as the credibility parameter; Step 21101: When the transmission data information is obtained, the fault signal is defined as a data integrity abnormality signal; Step 211010: Obtain complete data information from the preset backup line; Step 211011: Obtaining a data loss reason from a preset data loss principle table based on the complete data information and the transmitted data information; Step 211012: Obtain integrity correction parameters from a preset data integrity parameter table based on the data loss cause and the suspected fault cause; Step 211013: Obtain an integrity correction credibility parameter based on the credibility parameter and the integrity correction parameter, and output it as the credibility parameter.

7. A HPLC fault diagnosis method according to claim 6, characterized in that, The method for obtaining and outputting the credibility parameter also includes: Step 21102: when the data integrity abnormality signal and the data acquisition abnormality signal are received alternately during a preset historical period, defining the fault signal as a mixed abnormality signal; Step 211020: Analyze the proportion of the data integrity abnormal signal and the data acquisition abnormal signal in the mixed abnormal signal in the historical time, and define it as the mixed abnormality proportion; Step 211021: Simulate fault results based on the suspected fault causes corresponding to all the fault analysis solutions to obtain a simulated abnormal signal; Step 211022: Analyze the proportion of the data integrity abnormal signal and the data acquisition abnormal signal in the simulated abnormal signal in the historical time, and define it as the simulated abnormal proportion; Step 211023: obtaining a mixing correction parameter from a preset mixing correction database according to the ratio of the mixing anomaly proportion to the simulated anomaly proportion; Step 211024: Obtain a mixed correction credibility parameter based on the credibility parameter and the mixed correction parameter, and continue to execute step 212 using the mixed correction credibility parameter as the credibility parameter.

8. A HPLC fault diagnosis method according to claim 1, characterized in that, A diagnostic method when the credibility parameter is lower than the credibility threshold, the method comprising: Step 2120: Sort the suspected fault causes based on the fault probabilities to obtain a suspected fault cause ranking table; Step 2121: traverse the suspected fault cause ranking table, and define the suspected fault cause next to the suspected fault cause as the next suspected fault cause; Step 21210: When the credibility parameter of the fault analysis solution corresponding to the suspected fault cause is lower than the credibility threshold, the second suspected fault cause is taken as the suspected fault cause and the process continues with step 2121; Step 21211: When the credibility parameter of the fault analysis solution corresponding to the suspected fault cause is higher than the credibility threshold, output the suspected fault cause corresponding to the credibility parameter.

9. A HPLC fault diagnosis method according to claim 8, characterized in that, A diagnostic method when the credibility parameter of the fault analysis solution corresponding to the absence of the suspected fault cause is higher than the credibility threshold, the method comprising: Step 212100: Find all fault device points preset in the fault analysis plan; Step 212101: Obtain a fault analysis solution group based on integrating the fault analysis solutions corresponding to the faulty device points; Step 212102: Accumulate all the credibility parameters of the fault analysis solution group to obtain the device point credibility; Step 212103: When the device point credibility exceeds a preset device credibility threshold, obtaining a possible fault cause of the faulty device point corresponding to the device point credibility; Step 212104: Output the possible fault cause as the suspected fault cause.

10. A HPLC fault diagnosis method according to claim 1, characterized in that, A method for maintaining the fault analysis solution library, the method comprising: Step 3: After receiving the preset fault repair signal, obtain the fault cause and fault phenomenon corresponding to the fault signal; Step 30: Input the fault signal into the fault analysis solution library to obtain a corresponding fault ID; Step 31: Input the fault cause, the fault phenomenon, and the historical data information corresponding to the fault phenomenon into the fault analysis solution library according to the fault ID.

Images