Digital intelligent equipment supervision method and system based on data analysis algorithm

Abstract

The present application relates to the technical field of electronic equipment supervision, and specifically to a digital intelligent equipment supervision method and system based on a data analysis algorithm, which specifically includes: obtaining the representative characteristic value of the mutation of each electronic device based on the abnormality and acquisition time of the abnormal data in the current sequence and voltage sequence of each electronic device; constructing the propagation confidence between any two electronic devices based on the difference between the abnormal data of different electronic devices; obtaining the electrical interconnection index of each electronic device based on the correlation between the current sequences of different electronic devices and the correlation between the voltage sequences, combining the propagation confidence and the representative characteristic value of the mutation; and performing fault detection on each electronic device based on the electrical interconnection index and the differential sequence of the current sequence and the voltage sequence. This realizes digital intelligent equipment supervision, and improves the stability and reliability of the factory production line by analyzing the interconnectivity of electronic equipment faults.

Description

Technical Field

[0001] The present application relates to the technical field of electronic device supervision, and specifically to a digital intelligent device supervision method and system based on a data analysis algorithm. Background Art

[0002] The digital intelligent equipment supervision method and system based on data analysis algorithm integrates a comprehensive solution of multiple advanced technologies, aiming to achieve accurate monitoring and management of the status of electronic equipment through real-time data collection and in-depth analysis. As one of the core functions of the system, electronic equipment fault detection can predict and identify potential failure modes by analyzing the data generated during equipment operation, thereby achieving preventive maintenance. It not only improves the reliability and efficiency of equipment, but also significantly reduces maintenance costs and downtime, which is crucial to ensure the continuous operation of critical infrastructure and industrial applications. In addition, with the development of technology and the increase in equipment complexity, the importance of electronic equipment fault detection has become increasingly prominent. It is not only related to the economic efficiency of enterprise operations, but also involves personnel safety and social stability. Therefore, building an efficient and intelligent supervision system is of great significance for extending equipment life, optimizing resource allocation, improving decision-making quality, and enhancing enterprise competitiveness.

[0003] In modern production plants, the connectivity of electronic devices and machines is key to ensuring that production processes run efficiently. However, this high degree of connectivity also brings potential risks, namely that the failure of one device may trigger a chain reaction that affects the entire production line. The dependencies between devices mean that the failure of one component will quickly spread to other components, and the series of reactions caused by the failure may cause the performance of the entire system to deteriorate. Traditional fault detection methods can often only identify the failure of a single device, but cannot identify how the failure propagates between devices. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a digital intelligent device supervision method and system based on data analysis algorithm. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a digital intelligent device supervision method based on a data analysis algorithm, the method comprising the following steps:

[0006] Collect the current and voltage data of each electronic device at each moment, and perform preprocessing to construct the current sequence and voltage sequence of each electronic device;

[0007] The abnormal values ​​in the current sequence and the voltage sequence and the abnormality degree of each abnormal value are respectively obtained by an abnormal detection algorithm; each synchronous abnormal value group of each electronic device is obtained based on the difference in the acquisition time of the abnormal values ​​in the current sequence and the voltage sequence; the synchronous abnormal change time is determined based on the acquisition time of two abnormal values ​​in the synchronous abnormal value group;

[0008] Based on the size of the outliers in the synchronous outlier group and the difference between the corresponding acquisition times, combined with the degree of disorder of the synchronous mutation time of the electronic device, the representative characteristic value of the mutation of each electronic device is constructed;

[0009] Based on the difference between the abnormal data of different electronic devices, the propagation confidence between any two electronic devices is constructed; based on the correlation between the current sequences of different electronic devices and the correlation between the voltage sequences, the electrical interconnection index of each electronic device is obtained by combining the propagation confidence and the representative characteristic value of the mutation;

[0010] According to the electrical interconnection index, the fault detection of each electronic device is carried out in combination with the differential sequence of the current sequence and the voltage sequence.

[0011] In one embodiment, the process of acquiring each synchronization abnormal value group of each electronic device is as follows:

[0012] The abnormal values ​​in the current series are recorded as current abnormal values, and the abnormal values ​​in the voltage series are recorded as voltage abnormal values;

[0013] The difference between the collection time of any current abnormal value and the collection time of any voltage abnormal value of the same electronic device is calculated. If the difference is less than the time threshold, the combination of any current abnormal value and any voltage abnormal value is recorded as a synchronous abnormal value group.

[0014] In one embodiment, the synchronous abnormal change time is: the collection time corresponding to the average of the collection times of two abnormal values ​​in each synchronous abnormal value group is recorded as the synchronous abnormal change time.

[0015] In one embodiment, the process of acquiring the mutation representative feature value of each electronic device is as follows:

[0016] Calculate the chaos degree of all synchronous abnormality moments of a single electronic device, recorded as the first chaos degree; record the mean of the abnormality degrees of the two abnormal values ​​in each synchronous abnormal value group as the first mean of abnormality; record the difference between the collection moments of the two abnormal values ​​in each synchronous abnormal value group as the first collection moment difference; record the forward fusion result of the first abnormality mean and the first collection moment difference of each synchronous abnormal value group as the first fusion value; the representative characteristic value of the mutation of the electronic device is positively correlated with the first chaos degree, and negatively correlated with the first fusion value.

[0017] In one embodiment, the process of acquiring the propagation confidence between any two electronic devices is as follows:

[0018] Calculate the difference in the number of synchronization abnormal value groups between any two electronic devices, and record it as the first number difference; calculate the difference in the first disorder degree between any two electronic devices, and record it as the first disorder degree difference;

[0019] The propagation confidence between any two electronic devices is positively correlated with the first numerical difference and the first chaos difference, respectively.

[0020] In one embodiment, the process of obtaining the electrical appliance interconnection index of each electronic device is as follows:

[0021] Obtaining electrical correlation between any two electronic devices based on correlation between current sequences and correlation between voltage sequences of different electronic devices;

[0022] Calculate the average value of the electrical correlation between any electronic device and all other electronic devices, recorded as the first average value; calculate the average value of the propagation confidence between any electronic device and all other electronic devices, recorded as the second average value; calculate the ratio between the first average value and the second average value of any electronic device, recorded as the first ratio; the electrical interconnection index of any electronic device is positively correlated with the first ratio and the representative characteristic value of the mutation of any electronic device, respectively.

[0023] In one embodiment, the process of obtaining the appliance correlation between any two electronic devices is as follows:

[0024] The correlation between the current sequences of any two electronic devices is calculated and recorded as current correlation; the correlation between the voltage sequences of any two electronic devices is calculated and recorded as voltage correlation; the average value between the current correlation and the voltage correlation of any two electronic devices is taken as the electrical correlation of the any two electronic devices.

[0025] In one embodiment, the fault detection of each electronic device is performed based on the electrical interconnection index in combination with the differential sequence of the current sequence and the voltage sequence, specifically:

[0026] Calculate the difference sequence of the current sequence and voltage sequence of each electronic device;

[0027] Each differential data greater than the jump threshold in the current differential sequence of any electronic device is recorded as current fault differential data; each differential data greater than the jump threshold in the voltage differential sequence of each electronic device is recorded as voltage fault differential data; two acquisition moments corresponding to each current fault differential data are recorded as a current fault moment, and two acquisition moments corresponding to each voltage fault differential data are recorded as a voltage fault moment; all current fault moments and all voltage fault moments of any electronic device are recorded as fault moments;

[0028] At each fault moment of any of the electronic devices, obtaining a maximum value among the current differential data and the voltage differential data of all remaining electronic devices except the any of the electronic devices, and recording it as the maximum value of the differential data at the fault moment;

[0029] The maximum value of the differential data at each fault moment of any electronic device is corrected, and based on the corrected maximum value of the differential data, it is analyzed whether any electronic device generates a fault association with other electronic devices at each fault moment.

[0030] In one embodiment, the maximum value of the differential data at each fault moment of any electronic device is corrected, and based on the corrected maximum value of the differential data, whether any electronic device generates a fault association with other electronic devices at each fault moment is analyzed, specifically:

[0031] Calculate the sum of the normalized value of the electrical interconnection index of any electronic device and the adjustment parameter, and record it as a first sum; multiply the first sum by the maximum value of the differential data at each fault moment of any electronic device as a correction value of the maximum value of the differential data at each fault moment of any electronic device;

[0032] If the correction value of the maximum value of the differential data at each fault moment of any of the electronic devices is greater than the jump threshold, it means that when any of the electronic devices fails, the electronic device corresponding to the maximum value of the differential data at the fault moment also fails due to fault interconnectivity; otherwise, no failure occurs due to fault correlation.

[0033] In the second aspect, an embodiment of the present application also provides a digital intelligent device supervision system based on a data analysis algorithm, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above methods are implemented.

[0034] The embodiments of the present application have at least the following beneficial effects:

[0035] The present application proposes a digital intelligent equipment supervision method and system based on a data analysis algorithm, which improves the stability and reliability of the factory production line by analyzing the interconnectivity of electronic equipment faults; for the rational anomaly of a single electronic device, the COF anomaly monitoring algorithm is used to identify the rational mutation of a single device to facilitate the subsequent correlation analysis between different devices; for the rational anomaly propagation problem between different electronic devices, the evaluation of the rational mutation propagation between devices is quantified through the differences in the synchronous anomaly value groups and synchronous abnormality moments of different devices; for the similarity of electrical characteristics and the possibility of fault propagation between electronic devices, the correlation of current and voltage sequences is analyzed through the Pearson correlation algorithm to solve the problem of quantitative evaluation of the mutual dependence between devices and the synchronization of fault propagation; when a faulty device occurs, it is judged whether the remaining devices will be affected by the fault correlation and cause faults based on the jump threshold of the current and voltage differential sequence and the correction method of the electrical interconnection index, so as to solve the problem of stability and fault prevention of the entire electronic equipment group. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 A flowchart of the steps of a digital intelligent device supervision method based on a data analysis algorithm provided in one embodiment of the present application;

[0038] Figure 2 This is a schematic diagram of the specific steps of the digital intelligent device supervision method based on data analysis algorithm. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, describes in detail the specific implementation methods, structures, features and effects of the digital intelligent device supervision method and system based on the data analysis algorithm proposed in the present application. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0041] The specific scheme of the digital intelligent device supervision method and system based on data analysis algorithm provided by the present application is described in detail below with reference to the accompanying drawings.

[0042] See also Figure 1 , which shows a flowchart of a digital intelligent device supervision method based on a data analysis algorithm provided by an embodiment of the present application, the method comprising the following steps:

[0043] Step S1, collecting current and voltage data of each electronic device at each moment, and performing preprocessing to construct a current sequence and a voltage sequence of each electronic device.

[0044] By installing current and voltage integrated sensors on all key electronic devices in the electronic equipment group on the production line, real-time monitoring of the two key parameters of equipment current and voltage can be achieved. The current and voltage integrated sensors transmit the collected data to the central processing unit through a wireless or wired network, and transmit the collected data to a user-friendly interface to display the real-time current and voltage data of different devices. It should be noted that users can add relevant information of the equipment in the user-friendly interface according to actual conditions, such as operating status, usage time, etc.

[0045] For all electronic devices that need to be monitored, when at least one device starts to run, the current and voltage data of all devices are collected at an interval of 0.2s. The current and voltage data of all devices are obtained in real time, and all the collected current and voltage data are normalized. The sequence composed of all the normalized values ​​of the current data of each electronic device in ascending time order is recorded as the current sequence of each electronic device, and the sequence composed of all the normalized values ​​of the voltage data of each electronic device in ascending time order is recorded as the voltage sequence of each electronic device.

[0046] Step S2, obtaining a representative characteristic value of a sudden change of each electronic device based on the abnormality degree and acquisition time of abnormal data in the current sequence and voltage sequence of each electronic device.

[0047] Before analyzing the interconnectedness of faults between electronic devices, analyzing a single device can help us understand how faults propagate between devices. For example, overheating of a device may cause its performance to degrade, which in turn affects the devices connected to it. Not only that, analyzing a single electronic device not only helps to understand the working principle and failure mode of a single device, but is also crucial to the stability, reliability and safety of the entire system.

[0048] Since the electronic equipment in the factory production line is relatively complex and may contain many nonlinear electronic components such as diodes, the current and voltage of each device are not simply linearly related. And due to load changes and power fluctuations, the current and voltage of electronic equipment may undergo reasonable mutations. For example, when the electronic equipment is connected to different loads or the load demand changes, the current and voltage may be quickly adjusted to adapt to the new working conditions. When the equipment is started or shut down, the current and voltage may experience a short transition period, which is the natural process of the equipment switching from one state to another. The circuit may contain mechanisms such as overload protection and short-circuit protection, which quickly cut off or adjust the current and voltage when abnormal conditions are detected. In order to analyze the fault interconnection characteristics between different electronic devices, the circuit correlation between different electronic devices can be judged from the reasonable mutations of electronic devices.

[0049] (1) For a single electronic device, the current sequence and voltage sequence of the electronic device are respectively used as the input of the anomaly detection algorithm, and the output is the abnormal values ​​in the current sequence and voltage sequence and the abnormality degree of each abnormal value; the abnormal value in the current sequence is recorded as the current abnormal value, and the abnormal value in the voltage sequence is recorded as the voltage abnormal value.

[0050] It should be noted that there are many existing anomaly detection algorithms. Implementers can use anomaly detection algorithms such as isolation forest, local outlier factor algorithm (LOF), COF anomaly detection algorithm, etc. to perform anomaly detection on current series and voltage series. This embodiment uses the COF anomaly detection algorithm for anomaly detection.

[0051] (2) For the abnormal current value and abnormal voltage value of the same electronic device, the difference between the acquisition time of any abnormal current value and the acquisition time of any abnormal voltage value is calculated. If the difference is less than the time threshold, the combination of any abnormal current value and any abnormal voltage value is recorded as a synchronous abnormal value group. It should be noted that the time threshold is a preset value, which can be set by the implementer according to the actual situation. In this embodiment, the time threshold is set to 1s.

[0052] (3) Calculate the mean of the collection time of the two outliers in each synchronous outlier group, and record the collection time corresponding to the mean as the synchronous abnormal change time; calculate the chaos degree of all synchronous abnormal change times of a single electronic device, and record it as the first chaos degree;

[0053] The average of the abnormality of the two abnormal values ​​in each synchronous abnormal value group is recorded as the first abnormality average; the difference between the acquisition time of the two abnormal values ​​in each synchronous abnormal value group is recorded as the first acquisition time difference; the forward fusion result of the first abnormality average and the first acquisition time difference of each synchronous abnormal value group is recorded as the first fusion value;

[0054] The mutation representative characteristic value of the electronic device is positively correlated with the first chaos degree and negatively correlated with the first fusion value.

[0055] Preferably, as an embodiment of the present application, the first disorder degree may be the standard deviation of all synchronous abnormality moments of a single electronic device; the first fusion value may be the product of the first abnormality mean of each synchronous abnormality value group and the first acquisition moment difference; the sum of the first fusion values ​​of all synchronous abnormality value groups of a single electronic device is calculated; the mutation representative characteristic value of a single electronic device may be the ratio of the first disorder degree of the single electronic device to the sum of the first fusion values;

[0056] In other embodiments of the present application, the first degree of chaos may be the variance of all synchronous abnormal change moments of a single electronic device.

[0057] The first chaos degree analyzes the distribution characteristics of synchronous anomalies in a single electronic device. If the value of the first chaos degree is larger, it indicates that the device has experienced multiple independent abnormal changes at different time points, and there are multiple scattered potential problem points or unstable factors in the electronic device. At this time, when judging the circuit correlation between different electronic devices, the stronger the mutation representativeness of the electronic device, the larger the mutation representative characteristic value A of the electronic device; the first fusion value analyzes the mutation difference characteristics of the synchronous anomaly value group. If the value of the first fusion value is larger, it means that the difference between the current anomaly values ​​and the voltage anomaly values ​​in each synchronous anomaly value group is more significant, and the synchronization in time is worse. At this time, when judging the circuit correlation between different electronic devices, the mutation representativeness of the electronic device is weaker, and the mutation representative characteristic value A of the electronic device is smaller.

[0058] Step S3, constructing the propagation confidence between any two electronic devices based on the difference in the number of synchronous anomaly value groups of different electronic devices and the difference in the first chaos degree; based on the correlation between the current sequences of different electronic devices and the correlation between the voltage sequences, combined with the mutation representative characteristic value of each electronic device, obtaining the electrical interconnection index of each electronic device.

[0059] In the context of fault propagation between devices, the abnormal behavior of a single device may be affected by other devices or affect the status of other devices. In order to quantify this interaction, it is necessary to further explore the differences in the occurrence of reasonable mutations between two devices.

[0060] (1) Based on the difference in the number of synchronous outlier value groups of different electronic devices and the difference in the first chaos degree, the propagation confidence between any two electronic devices is constructed:

[0061] Calculate the difference in the number of synchronization outlier value groups between any two electronic devices, and record it as the first number difference; calculate the difference in the first disorder degree between any two electronic devices, and record it as the first disorder degree difference;

[0062] The propagation confidence between any two electronic devices is positively correlated with the first number difference and the first confusion difference between any two electronic devices.

[0063] Preferably, as an embodiment of the present application, the propagation confidence between any two electronic devices may be the product of the first numerical difference and the first chaos difference between any two electronic devices.

[0064] The larger the difference of the first number, the more it means that the time when the abnormalities of the two devices occur is not consistent, indicating that the mutual influence between them is weak, or their abnormalities are caused by different factors, and the larger the propagation confidence B; the first chaos degree difference analyzes the consistency of the time when the abnormalities of the two electronic devices occur. The larger the difference of the first chaos degree, it means that even if abnormalities occur in both devices, the specific time points of these abnormalities are quite different, which means that the synchronization of fault propagation is poor, and the propagation confidence B increases accordingly.

[0065] (2) Based on the correlation between the current sequences and the voltage sequences of different electronic devices, the electrical interconnection index of each electronic device is obtained by combining the propagation confidence and the representative characteristic value of the mutation:

[0066] When analyzing the fault interconnectivity between two electronic devices, in addition to considering the differences in the occurrence of reasonable mutations between them, it is also necessary to pay attention to the similarities in the current and voltage characteristics of the two devices. Current and voltage similarity analysis helps to identify potential correlations between devices. For example, if two devices exhibit similar current and voltage characteristics under the same operating conditions, then there is some form of mutual dependence between them.

[0067] Calculate the correlation between the current sequences of any two electronic devices, which is recorded as current correlation; similarly, calculate the correlation between the voltage sequences of any two electronic devices, which is recorded as voltage correlation; take the average value between the current correlation and the voltage correlation of any two electronic devices as the electrical correlation of any two electronic devices;

[0068] Calculate the average value of the electrical correlation between any electronic device and all other electronic devices, recorded as the first average value; calculate the average value of the propagation confidence between any electronic device and all other electronic devices, recorded as the second average value; calculate the ratio between the first average value and the second average value of any electronic device, recorded as the first ratio; the electrical interconnection index of any electronic device is positively correlated with the first ratio and the representative characteristic value of the mutation of any electronic device, respectively.

[0069] Preferably, as an embodiment of the present application, the correlation between the current sequences may be the Pearson correlation coefficient between the current sequences of any two electronic devices; the electrical interconnection index of any electronic device may be the product of the first ratio of any electronic device and the representative characteristic value of the mutation;

[0070] In other embodiments of the present application, the correlation between the current sequences may be the cosine similarity between the current sequences of any two electronic devices.

[0071] The interdependence between different devices and the possibility of fault propagation are quantified by the first ratio. The larger the first ratio is, the higher the similarity of the electrical characteristics between different devices is, the stronger the synchronization of fault propagation is, and the larger the electrical appliance interconnection index is. The representative characteristic value of the mutation of each electronic device reflects the frequency and dispersion of abnormal changes occurring in a single device at different time points, as well as the difference and time synchronization between abnormal values. The larger the value is, the stronger the mutation representativeness of a single device is, indicating that the device is more important or influential in the entire device group, and the electrical appliance interconnection index increases accordingly.

[0072] Step S4, performing fault detection on each electronic device according to the electrical interconnection index and the differential sequence of the current sequence and the voltage sequence.

[0073] Calculate the difference sequence of the current sequence and the voltage sequence of each electronic device; set a jump threshold. It should be noted that the jump threshold can be set by the implementer. In this embodiment, the jump threshold is set to 0.4;

[0074] Each differential data greater than the jump threshold in the current differential sequence of any electronic device is recorded as current fault differential data; each differential data greater than the jump threshold in the voltage differential sequence of each electronic device is recorded as voltage fault differential data; two acquisition moments corresponding to each current fault differential data are recorded as a current fault moment, and two acquisition moments corresponding to each voltage fault differential data are recorded as a voltage fault moment; all current fault moments and all voltage fault moments of any electronic device are recorded as fault moments;

[0075] At each fault moment of any of the electronic devices, obtaining a maximum value among the current differential data and the voltage differential data of all remaining electronic devices except the any of the electronic devices, and recording it as the maximum value of the differential data at the fault moment;

[0076] The maximum value of the differential data at each fault moment of any electronic device is corrected, and whether any electronic device generates a fault association with other electronic devices at each fault moment is analyzed based on the corrected maximum value of the differential data, specifically:

[0077] Set the adjustment parameter. The implementer can set the adjustment parameter by himself. In this embodiment, the adjustment parameter is set to 1.

[0078] Calculate the sum of the normalized value of the electrical interconnection index of any electronic device and the adjustment parameter, and record it as a first sum; multiply the first sum by the maximum value of the differential data at each fault moment of any electronic device as a correction value of the maximum value of the differential data at each fault moment of any electronic device;

[0079] If the correction value of the maximum value of the differential data at each fault moment of any of the electronic devices is greater than the jump threshold, it means that when any of the electronic devices fails, the electronic device corresponding to the maximum value of the differential data at the fault moment also fails due to fault interconnectivity; otherwise, no failure occurs due to fault correlation.

[0080] Finally, the equipment fault information caused by fault interconnectivity is displayed on the established user-friendly interface to facilitate maintenance personnel to quickly identify and respond.

[0081] The specific steps of the above method are shown in the schematic diagram Figure 2 shown.

[0082] Based on the same inventive concept as the above method, an embodiment of the present application also provides a digital intelligent device supervision system based on a data analysis algorithm, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned digital intelligent device supervision methods based on data analysis algorithms when executing the computer program.

[0083] In summary, the embodiments of the present application provide a digital intelligent device supervision method based on a data analysis algorithm, which improves the stability and reliability of a factory production line by analyzing the interconnectivity of electronic equipment faults; for the rational anomaly of a single electronic device, the COF anomaly monitoring algorithm is used to identify the rational mutation of a single device, so as to facilitate the subsequent correlation analysis between different devices; for the rational anomaly propagation problem between different electronic devices, the evaluation of the rational mutation propagation between devices is quantified through the differences in the synchronous anomaly value groups and synchronous abnormality moments of different devices; for the similarity of electrical characteristics and the possibility of fault propagation between electronic devices, the correlation of current and voltage sequences is analyzed through the Pearson correlation algorithm to solve the problem of quantitative evaluation of interdependence between devices and synchronization of fault propagation; when a faulty device occurs, it is determined whether the remaining devices will be affected by the fault correlation and cause faults based on the jump threshold of the current and voltage differential sequence and the correction method of the electrical interconnection index, so as to solve the problem of stability and fault prevention of the entire electronic equipment group.

[0084] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. The above-mentioned specific embodiments of the present application are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0086] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A digital intelligent device supervision method based on data analysis algorithm, characterized in that: The method comprises the following steps: Collect the current and voltage data of each electronic device at each moment, and perform preprocessing to construct the current sequence and voltage sequence of each electronic device; The abnormal values ​​in the current sequence and the voltage sequence and the abnormality degree of each abnormal value are respectively obtained by an abnormal detection algorithm; each synchronous abnormal value group of each electronic device is obtained based on the difference in the acquisition time of the abnormal values ​​in the current sequence and the voltage sequence; the synchronous abnormal change time is determined based on the acquisition time of two abnormal values ​​in the synchronous abnormal value group; Based on the size of the outliers in the synchronous outlier group and the difference between the corresponding acquisition times, combined with the degree of disorder of the synchronous mutation time of the electronic device, the representative characteristic value of the mutation of each electronic device is constructed; Based on the difference between the abnormal data of different electronic devices, the propagation confidence between any two electronic devices is constructed; based on the correlation between the current sequences of different electronic devices and the correlation between the voltage sequences, the electrical interconnection index of each electronic device is obtained by combining the propagation confidence and the representative characteristic value of the mutation; According to the electrical interconnection index, the fault detection of each electronic device is carried out in combination with the differential sequence of the current sequence and the voltage sequence; The process of obtaining the mutation representative characteristic value of each electronic device is as follows: Calculate the chaos degree of all synchronous abnormal mutation moments of a single electronic device, recorded as the first chaos degree; record the average of the abnormality degrees of the two abnormal values ​​in each synchronous abnormal value group as the first abnormality average; record the difference between the acquisition moments of the two abnormal values ​​in each synchronous abnormal value group as the first acquisition moment difference; record the forward fusion result of the first abnormality average and the first acquisition moment difference of each synchronous abnormal value group as the first fusion value; the representative characteristic value of the mutation of the electronic device is positively correlated with the first chaos degree, and negatively correlated with the first fusion value; The process of obtaining the propagation confidence between any two electronic devices is as follows: Calculate the difference in the number of synchronization abnormal value groups between any two electronic devices, and record it as the first number difference; calculate the difference in the first disorder degree between any two electronic devices, and record it as the first disorder degree difference; The propagation confidence between any two electronic devices is positively correlated with the first number difference and the first chaos difference respectively; The process of obtaining the electrical interconnection index of each electronic device is as follows: Obtaining electrical correlation between any two electronic devices based on correlation between current sequences and correlation between voltage sequences of different electronic devices; Calculate the average value of the electrical correlation between any electronic device and all other electronic devices, recorded as the first average value; calculate the average value of the propagation confidence between any electronic device and all other electronic devices, recorded as the second average value; calculate the ratio between the first average value and the second average value of any electronic device, recorded as the first ratio; the electrical interconnection index of any electronic device is positively correlated with the first ratio and the representative characteristic value of the mutation of any electronic device, respectively.

2. The digital intelligent device supervision method based on data analysis algorithm according to claim 1 is characterized in that: The acquisition process of each synchronous abnormal value group of each electronic device is as follows: The abnormal values ​​in the current series are recorded as current abnormal values, and the abnormal values ​​in the voltage series are recorded as voltage abnormal values; The difference between the collection time of any current abnormal value and the collection time of any voltage abnormal value of the same electronic device is calculated. If the difference is less than the time threshold, the combination of any current abnormal value and any voltage abnormal value is recorded as a synchronous abnormal value group.

3. The digital intelligent device supervision method based on data analysis algorithm according to claim 1 is characterized in that: The synchronous abnormal change time is: the collection time corresponding to the mean of the collection times of two abnormal values ​​in each synchronous abnormal value group is recorded as the synchronous abnormal change time.

4. The digital intelligent device supervision method based on data analysis algorithm according to claim 1 is characterized in that: The process of obtaining the electrical appliance correlation between any two electronic devices is as follows: The correlation between the current sequences of any two electronic devices is calculated and recorded as current correlation; the correlation between the voltage sequences of any two electronic devices is calculated and recorded as voltage correlation; the average value between the current correlation and the voltage correlation of any two electronic devices is taken as the electrical correlation of the any two electronic devices.

5. The digital intelligent device supervision method based on data analysis algorithm according to claim 1 is characterized in that: The method of performing fault detection on each electronic device according to the electrical interconnection index and in combination with the differential sequence of the current sequence and the voltage sequence is specifically as follows: calculating the differential sequence of the current sequence and the voltage sequence of each electronic device; Record each differential data greater than the jump threshold in the current differential sequence of any electronic device as current fault differential data; Record each differential data greater than the transition threshold in the voltage differential sequence of each electronic device as voltage fault differential data; Recording the two acquisition moments corresponding to each current fault differential data as a current fault moment, and recording the two acquisition moments corresponding to each voltage fault differential data as a voltage fault moment; and taking all current fault moments and all voltage fault moments of any electronic device as fault moments; At each fault moment of any of the electronic devices, obtaining a maximum value among the current differential data and the voltage differential data of all remaining electronic devices except the any of the electronic devices, and recording it as the maximum value of the differential data at the fault moment; The maximum value of the differential data at each fault moment of any electronic device is corrected, and based on the corrected maximum value of the differential data, it is analyzed whether any electronic device generates fault interconnection with other electronic devices at each fault moment.

6. The digital intelligent device supervision method based on data analysis algorithm according to claim 5 is characterized in that: The step of correcting the maximum value of the differential data at each fault moment of any electronic device, and analyzing whether any electronic device generates fault interconnection with other electronic devices at each fault moment based on the corrected maximum value of the differential data, is specifically as follows: Calculate the sum of the normalized value of the electrical interconnection index of any electronic device and the adjustment parameter, and record it as a first sum; multiply the first sum by the maximum value of the differential data at each fault moment of any electronic device as a correction value of the maximum value of the differential data at each fault moment of any electronic device; If the correction value of the maximum value of the differential data at each fault moment of any of the electronic devices is greater than the jump threshold, it means that when any of the electronic devices fails, the electronic device corresponding to the maximum value of the differential data at the fault moment also fails due to fault interconnectivity; otherwise, no failure occurs due to fault interconnectivity.

7. A digital intelligent device supervision system based on a data analysis algorithm, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

Images