Method and system for monitoring abnormal working condition electricity utilization information

Abstract

The invention discloses a working condition electricity utilization information abnormity monitoring method and system, and relates to the technical field of power monitoring. According to the method, the abnormal type judgment process is constructed, when the abnormal data points are detected, the production areas where the abnormal data points are located are identified firstly, different judgment steps are executed according to whether the abnormal data points are power consumption data or power anomalies and the number of the abnormal data points, the abnormal types are determined in combination with historical operation and maintenance records, and therefore abnormal type judgment is more accurate and more targeted; the problems that in the prior art, power consumption information can only be monitored, the power consumption abnormity type cannot be judged when abnormity is monitored, and when power consumption abnormity occurs, abnormity can only be roughly judged, but the specific abnormity type is difficult to clarify are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring, and particularly to a method and system for abnormal monitoring of working condition power consumption information. Background Art

[0002] With the rapid development of industry, the number of enterprise power consumption equipment is increasing continuously, and the power consumption situation is becoming increasingly complex. Timely and accurately grasping the working condition information of power consumption equipment is of great significance for ensuring the safe and stable operation of the power system, improving energy utilization efficiency, and achieving the goals of energy conservation and emission reduction.

[0003] However, in the actual application process of the existing methods for abnormal monitoring of working condition power consumption information, there are still the following deficiencies: The existing monitoring means have limitations in data collection, and it is difficult to obtain power consumption data comprehensively and accurately. On the one hand, the data collection points are not set comprehensively enough to cover the key power consumption equipment in each production area of the enterprise, resulting in the lack of some power consumption information and being unable to reflect the overall power consumption situation. In addition, most of them can only monitor the power consumption information and cannot judge the type of power consumption abnormality when an abnormality is monitored. When a power consumption abnormality occurs, it is often only possible to roughly judge that there is an abnormality, but it is difficult to clarify the specific type of abnormality. This makes the subsequent treatment measures lack pertinence, unable to solve the power consumption abnormality problem in a timely and effective manner, and increasing the risk of the operation of the power system.

[0004] Therefore, a method and system for abnormal monitoring of working condition power consumption information are introduced. Summary of the Invention

[0005] In view of this, the present invention provides a method and system for abnormal monitoring of working condition power consumption information to solve the problems raised in the above background art.

[0006] The object of the present invention can be achieved by the following technical solutions: A method for abnormal monitoring of working condition power consumption information includes: Establishing a power consumption model: presetting the update time interval of the normal power consumption range, and after reaching the update time interval, testing the change range of the power consumption data in each production area under different working conditions and the change range of the power of each power consumption equipment as the normal power consumption range and the normal power range, and inputting them into the pre-constructed power consumption model for updating; Real-time data comparison: for the power consumption data of different production areas and the power data of each power consumption equipment in different production areas, taking them as each group of data points and inputting them into the power consumption model to compare with the corresponding normal power consumption range or normal power range. If a certain group of data points is higher than the highest value within the corresponding normal power consumption range or normal power range, it is marked as an abnormal data point.

[0007] Abnormal type judgment: When abnormal data points are detected at a certain moment, after identifying the production area where the abnormal data points are located, corresponding steps are executed in combination with historical data to initially judge the abnormal type of the abnormal data points.

[0008] In some embodiments, after identifying the production area where the abnormal data points are located, corresponding steps are executed in combination with historical data, specifically as follows: : The abnormal data points corresponding to the electricity consumption data and power are represented by numbers a and b respectively; : If there is only one set of abnormal data points in the production area at the current moment, after identifying the numbers of the abnormal data points, corresponding steps or ; : If there are both abnormal data points a and b in the production area at the current moment, first count the number of abnormal data points b in the production area at the current moment. If the number is one, it is determined as a single electrical equipment abnormality. If the number is greater than one, it is determined as multiple electrical equipment abnormalities; For the combinations of single electrical equipment abnormality and multiple electrical equipment abnormalities with abnormal data a, corresponding steps or .

[0009] In some embodiments, when executing step to initially judge the abnormal type of the abnormal data points, specifically as follows: : If there is only abnormal data point a in the production area at the current moment, match the current moment with the working time range of the corresponding production area. If the match fails, the abnormal type is initially judged as illegal electricity consumption. Otherwise, extract the abnormal reasons for the recent x times (where x > 5) when there is only abnormal data point a within the working time range from the historical data of the corresponding production area; use the abnormal reason with the highest occurrence frequency as the initially judged abnormal type.

[0010] In some embodiments, when executing step to initially judge the abnormal type of the abnormal data points, specifically as follows: : If there is only abnormal data point b in the production area at the current moment, further identify the electrical equipment corresponding to abnormal data point b, and then extract the highest value within the normal power range of the corresponding electrical equipment from the electricity consumption model; use the highest value within the normal power range as the power threshold of the corresponding electrical equipment, and observe the change of abnormal data b of the corresponding electrical equipment within the set time window after the current moment; If the continuous duration of the abnormal data b of the corresponding user equipment exceeding the power threshold within the set time window exceeds the preset reference duration, the abnormal type is initially judged as electrical equipment overload; If the duration for which the abnormal data b of the corresponding user device is higher than the power threshold within the set time window is lower than the preset reference duration, then extract the power at each time point within the set time window according to the pre-divided time intervals, and substitute it into the standard deviation formula for calculation to obtain the fluctuation value of the abnormal data b of the corresponding electrical device within the set time window; compare the fluctuation value of the abnormal data b with the preset fluctuation threshold. If the fluctuation value of the abnormal data b is higher than the fluctuation threshold, it is determined that there is a fault inside the device, and at the same time, retrieve the historical operation and maintenance records of the corresponding electrical device; the historical operation and maintenance records include the change situation of each historical abnormal data point b of the corresponding electrical device, the maintenance time, the maintenance personnel, and the abnormal reasons, and locate the abnormal type of the corresponding electrical device at the current moment based on the historical operation and maintenance records.

[0011] In some embodiments, locating the abnormal type of the corresponding electrical device at the current moment based on the historical operation and maintenance records specifically includes: Extract the power at each time point within the set time window according to the pre-divided time intervals, and construct the current power sequence, denoted as ; is the power at each time point within the set time window, and n is the number of sampling points within the set time window; Screen the change situations of each historical abnormal data b of the corresponding electrical device in the historical operation and maintenance records, retain the records where the fluctuation value of the abnormal data b of the corresponding electrical device is higher than the fluctuation threshold, and extract the power sequence of each abnormal data b within the set time window from the retained records, denoted as ; where k is the number of each group of retained records, is the power value at the m-th time point in the k-th group of retained records; For the current power sequence Y and the power sequences of each historical group of retained records, calculate the similarity value between the power sequence of each historical group of retained records and the current power sequence Y using the cosine similarity; After obtaining the similarity values between the current power sequence Y and the power sequences of each historical group of retained records, generate a sorted list of the similarity values in descending order; Select the top three historical records with the highest similarity, and extract their abnormal reasons and historical occurrence frequencies. The historical occurrence frequency is the occurrence frequency of the corresponding abnormal reason in the historical operation and maintenance records; combine the similarity weights and the historical occurrence frequencies, calculate the confidence levels of the three groups of abnormal reasons. After normalizing the similarity and the historical occurrence frequencies, multiply the historical occurrence frequency and the similarity by the corresponding set weights respectively, and divide the result of multiplying the historical occurrence frequency by the corresponding weight by the result of multiplying the similarity by the corresponding weight to obtain the confidence levels of the three groups of abnormal reasons; Select the highest confidence abnormal cause from the three groups of abnormal causes as the initially judged abnormal type of the corresponding power-consuming device.

[0012] In some embodiments, perform the steps Initially judge the abnormal type of the abnormal data point, specifically: : For the abnormal situation of a single power-consuming device and abnormal data point a, after using the steps to identify the abnormal type of the corresponding power-consuming device, initially judge that the abnormal type is the abnormal power consumption data caused by the abnormal type of the corresponding power-consuming device.

[0013] In some embodiments, perform the steps Initially judge the abnormal type of the abnormal data point, specifically: : For the abnormal situation of multiple power-consuming devices and abnormal data point a, first check the input voltage of the distribution box in the production area at the current moment. If the measured input voltage is higher than the rated voltage, initially judge that the abnormal type is a public system abnormality; otherwise, After identifying the abnormal types of each abnormal power-consuming device, combine the fault data sets and initially judge that the abnormal type is the abnormal power consumption data caused by the fault data set.

[0014] In some embodiments, it further includes: Multi-point metering layout: Real-time collect the power consumption data of different production areas and the power of each power-consuming device in different production areas; Abnormal warning processing: After judging the abnormal type, trigger a warning signal and send it to the management personnel.

[0015] A working condition power consumption information abnormal monitoring system includes: Power consumption information collection module: Real-time collect the power consumption data of different production areas and the power of each power-consuming device in different production areas; Power consumption model update module: Preset the update time interval of the normal power consumption range. After reaching the update time interval, test the change range of the power consumption data of each production area under different working conditions and the change range of the power of each power-consuming device as the normal power consumption range and normal power range, and input them into the pre-constructed power consumption model for update; Abnormal evaluation module: Use the power consumption data of different production areas and the power data of each power-consuming device in different production areas as each group of data points and input them into the power consumption model to compare with the corresponding normal power consumption range or normal power range. If a certain group of data points is higher than the highest value within the corresponding normal power consumption range or normal power range, mark it as an abnormal data point; Abnormal type judgment module: When abnormal data points are detected at a certain moment, after identifying the production area where the abnormal data points are located, corresponding steps are executed in combination with historical data to preliminarily judge the abnormal type of the abnormal data points; Abnormal result transmission module: After judging the abnormal type, trigger a warning signal and send it to the management personnel Compared with the prior art, the beneficial effects of the present invention are: Through the constructed abnormal type judgment process, when the present invention detects abnormal data points, it first identifies the production area where they are located, and according to whether the abnormal data points are power consumption data or power abnormalities, and the number of abnormal data points, different judgment steps are executed. Combining historical operation and maintenance records to determine the abnormal type makes the judgment of the abnormal type more accurate and targeted, and solves the problem that most of the prior art can only monitor power consumption information and cannot judge the abnormal type of power consumption when monitoring abnormalities. When power consumption abnormalities occur, it is often only possible to roughly judge that there are abnormalities, but it is difficult to clarify the specific abnormal type; By presetting the update time interval of the normal power consumption range, after reaching the interval, the power consumption data and power change ranges under different working conditions in each production area are tested, and the pre-constructed power consumption model is updated, so that the model can describe the normal change ranges and mutual relationships of power consumption parameters under different production working conditions, adapt to the changes in enterprise production processes and equipment aging, and ensure the effectiveness and accuracy of the monitoring system; Through multi-point metering layout, sensors and electricity meters are installed at key power-consuming equipment, inlets and outlets in different production areas to achieve comprehensive and real-time collection of power consumption data. Combining data cleaning and normalization preprocessing improves the data quality and lays a foundation for accurately judging abnormalities. At the same time, a power consumption model is constructed and updated regularly. Based on historical data and real-time comparison, abnormal data points can be identified more accurately, solving the problem of inaccurate monitoring. Description of the Drawings

[0016] In the following description of the exemplary embodiments in conjunction with the drawings, more details, features and advantages of the present application are disclosed. In the drawings: Figure 1 is the flowchart of the present invention; Figure 2 is the principle block diagram of the present invention. Detailed Embodiments

[0017] The following will describe several embodiments of the present application in more detail with reference to the drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The described embodiments do not limit the present application.

[0018] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0019] Example 1 See also Figure 1 As shown, a method for monitoring abnormal power consumption information under working conditions includes: Multi-point metering layout: Electricity metering devices are installed on key electrical equipment in different industrial production areas; the electricity metering devices are sensors and meters, such as current sensors and power sensors installed on each electrical equipment, and smart meters installed at the inlet and outlet of the production area to collect electricity consumption data of different production areas and the power of each electrical equipment in different production areas in real time; to achieve multi-point metering of electricity consumption information; Pre-process the power consumption data of different production areas and the operation data of each power-consuming equipment in different production areas, including: Data cleaning: Clean the collected electricity consumption data to remove noise, outliers and missing values. Statistical analysis methods, such as standard deviation-based methods, can be used to identify and remove data points that deviate from the normal range; for missing values, interpolation methods or prediction methods based on historical data can be used to fill them; Data normalization: Normalize the electricity consumption data collected from different metering points to make them have the same dimension and range for subsequent analysis and comparison. The minimum-maximum normalization method or the z-score normalization method can be used; Establishing a power consumption model: preset the update time interval of the normal power consumption range. After the update time interval is reached, test the change range of power consumption data in each production area under different working conditions and the power change range of each power-consuming equipment as the normal power consumption range and normal power range, and input them into the pre-built power consumption model for updating; The normal power consumption range is completed by extracting the lowest and highest values from the variation range of power consumption data under different working conditions and combining them; the normal power range is the same; The model can describe the normal range of changes and interrelationships of power consumption parameters under different production conditions; the power consumption model can be updated regularly to adapt to changes in the company's production process and aging of equipment; incremental learning methods can be used to continuously add new power consumption data to the model for updating; Real-time data comparison: For the electricity consumption data in different production areas and the power data of each electrical equipment in different production areas, each group of data points is input into the electricity consumption model and compared with the corresponding normal electricity consumption range or normal power range. If a group of data points is higher than the highest value within the corresponding normal electricity consumption range or normal power range, it is marked as an abnormal data point; Abnormal type judgment: When abnormal data points are detected at a certain moment, after identifying the production area where the abnormal data points are located, corresponding steps are executed in combination with historical data to initially judge the abnormal type of the abnormal data points; Specifically: The abnormal data points corresponding to the electricity consumption data and power are represented by the numbers a and b respectively; : If there is only one group of abnormal data points in the production area at the current moment, after identifying the number of the abnormal data points, corresponding steps or ; : If there is only abnormal data point a in the production area at the current moment, match the current moment with the working time range of the corresponding production area. If the match fails, the abnormal type is initially judged as illegal electricity consumption. Otherwise, extract the abnormal reasons for the abnormal data point a that only exists within the working time range in the historical data of the corresponding production area for the last x times, where x > 5 and the specific value is set by the technical staff; the abnormal reasons are such as meter failure, inaccurate data transmission, etc.; the abnormal reason with the highest frequency of occurrence is used as the initially judged abnormal type; Example illustration, (1) Match the current moment with the working time range of the corresponding production area. For example, the normal working time of a production area is 8:00 - 18:00. If the current moment is 20:00 and the match fails, the abnormal type is initially judged as illegal electricity consumption because the abnormal electricity consumption data appears during non-working hours, which is likely to be unauthorized electricity consumption behavior; (2) Among the cases where only the electricity consumption data is abnormal within the last 10 working hours, the meter failure occurred 6 times. Then the abnormal type of this time is initially judged as meter failure; : If there is only abnormal data point b in the production area at the current moment, further identify the electrical equipment corresponding to abnormal data point b, and then extract the highest value within the normal power range of the corresponding electrical equipment from the electricity consumption model; use the highest value within the normal power range as the power threshold of the corresponding electrical equipment, and observe the change of the abnormal data b of the corresponding electrical equipment within the set time window after the current moment; If the continuous duration of the abnormal data b of the corresponding user equipment within the set time window is higher than the power threshold exceeds the preset reference duration, the abnormal type is initially judged as electrical equipment overload; If the duration for which the abnormal data b of the corresponding user device is higher than the power threshold within the set time window is lower than the preset reference duration, the power at each time point within the set time window is extracted according to the pre-divided time interval and substituted into the standard deviation formula for calculation to obtain the fluctuation value of the abnormal data b of the corresponding electrical device within the set time window; the fluctuation value of the abnormal data b is compared with the preset fluctuation threshold. If the fluctuation value of the abnormal data b is higher than the fluctuation threshold, it is determined that there is a fault inside the device, and at the same time, the historical operation and maintenance records of the corresponding electrical device are retrieved; the historical operation and maintenance records include the change situation, maintenance time, maintenance personnel, and abnormal reasons of each historical abnormal data point b of the corresponding electrical device. Extract the power at each time point within the set time window according to the pre-divided time interval, and construct the current power sequence, denoted as ; is the power at each time point within the set time window, and n is the number of sampling points within the set time window; Screen the change situation of each historical abnormal data b of the corresponding electrical device in the historical operation and maintenance records, retain the records where the fluctuation value of the abnormal data b of the corresponding electrical device is higher than the fluctuation threshold, and extract the power sequence of each abnormal data b within the set time window from the retained records, denoted as ; where k is the number of each group of retained records, is the power value at the m-th time point in the k-th group of retained records; For the current power sequence Y and the power sequences of each historical group of retained records, after normalization; use the cosine similarity to calculate the similarity value between the power sequences of each historical group of retained records and the current power sequence Y; Supplementary note, the similarity value calculation formula is expressed as ; is the similarity value; and n = m; After obtaining the similarity values between the current power sequence Y and the power sequences of each historical group of retained records, generate a sorted list of the similarity values of each group in descending order; Select the top three historical records with the highest similarity, extract their abnormal reasons and historical occurrence frequencies. The historical occurrence frequency is the occurrence frequency of the corresponding abnormal reason in the historical operation and maintenance records; combine the similarity weight and the historical occurrence frequency, calculate the confidence levels of the three groups of abnormal reasons. After normalizing the similarity and the historical occurrence frequency, multiply the historical occurrence frequency and the similarity by their corresponding set weights respectively, and divide the result of multiplying the historical occurrence frequency by the corresponding weight by the result of multiplying the similarity by the corresponding weight to obtain the confidence levels of the three groups of abnormal reasons; Supplementary note, the confidence level calculation formula is expressed as ; where is the confidence level, respectively representing the historical occurrence frequency and similarity, are the weights corresponding to the historical occurrence frequency and similarity respectively; Select the highest-confidence abnormal cause from the confidence levels of the three groups of abnormal causes as the initially judged abnormal type of the corresponding power-consuming equipment; The embodiment shows that (1) if the power corresponding to the abnormal data point is much higher than the highest value within the normal power consumption range and lasts for a long time, it may be that the equipment is overloaded. For example, if the normal power range of a certain motor is 10 - 20kW and the current power reaches 30kW and lasts for a period of time, it is very likely that the load driven by the motor is too large, then the initially judged abnormal type is overload; (2) If the power abnormally increases but fluctuates greatly, showing a sudden high and low power, it may be that there is a fault inside the equipment, such as a short circuit in the motor winding or damage to electrical components, and further combine historical operation and maintenance data to determine the cause of the fault; Supplementary explanation, after identifying the power-consuming equipment corresponding to the abnormal data point b, first combine the production process situation in the production area. If the production process has been adjusted recently, and the adjustment result is to add a process to the corresponding power-consuming equipment, then regard the production process change as the initially judged abnormal type of the abnormal data point. For example, adding a heating process in the production process may cause the power of the relevant equipment to increase; : If there are abnormal data points a and b in the production area at the current moment, first count the number of abnormal data points b in the production area at the current moment. If the number is one, it is determined as a single power-consuming equipment anomaly. If the number is greater than one, it is determined as a multi-power-consuming equipment anomaly; For the combination of single power-consuming equipment anomaly and multi-power-consuming equipment anomaly with abnormal data a, corresponding steps or ; If only the power and power consumption data of a certain piece of equipment are abnormal, it may be that the abnormal power consumption data is caused by a fault of the equipment itself; If multiple pieces of equipment are abnormal at the same time, it may be that there is a problem with the power supply system in the production area. For example, unstable power supply voltage will affect the operation of multiple pieces of equipment. Check the cause of the fault when multiple pieces of equipment are abnormal in the historical data, such as whether it is caused by a power supply system fault; : For single power-consuming equipment anomaly and abnormal data point a, use step After identifying the abnormal type of the corresponding power-consuming equipment, initially judge that the abnormal type is the abnormal power consumption data caused by the abnormal type of the corresponding power-consuming equipment; : For the anomalies of multiple electrical equipment and the anomaly data points a, first check the input voltage of the distribution box in the production area at the current moment. If the measured input voltage is higher than the rated voltage, preliminarily judge that the anomaly type is a common system anomaly; otherwise, go to step After identifying the anomaly types of each abnormal electrical equipment and combining the fault data sets, preliminarily judge that the anomaly type is the abnormal power consumption data caused by the fault data sets; Anomaly early warning processing: After judging the anomaly type, trigger an early warning signal and send it to the management personnel; After receiving the early warning signal, the management personnel notify the corresponding maintenance personnel to go to the production area on-site for inspection and maintenance; Supplementary description: Set the evaluation time interval for each production area. After reaching the set evaluation time interval, extract the historical data of anomaly monitoring in each production area within the current evaluation time window, and extract the preliminary judgment results and the on-site inspection results of the corresponding anomaly types of each anomaly data point in each production area within the current evaluation time window from the historical data; After the maintenance personnel repair the fault, manually enter the inspection results through the system interface or automatically synchronize them through the API to connect to the enterprise operation and maintenance management system (such as ERP, MES); Match the preliminary judgment results of each anomaly data point in each production area with the on-site inspection results. If the match is successful, the number of correct judgments is incremented by one; otherwise, the number of failed judgments is incremented by one; For the number of correct judgments in each production area, calculate its proportion in the total number of anomaly data points within the current evaluation time window, which is denoted as the accuracy rate; Extract the anomaly types of each correct judgment in each production area, and count the number of times of the actual anomaly types that occurred in each production area within the current evaluation time window, which is denoted as the actual occurrence times; For each anomaly type in each production area, calculate the proportion of the correct judgment parameter in the actual occurrence times, which is denoted as the recall rate. For the recall rate of each anomaly type corresponding to each production area, it is represented by Dc, where c is the number of each anomaly type, c = 1, 2,..., e, and e is the total number of anomaly types in each production area; Use the formula Calculate the anomaly monitoring status index Dra of each production area within the current evaluation time window; where V represents the accuracy rate of each production area within the current evaluation time window, and are the weights of the accuracy rate and the recall rate of each anomaly type respectively; Push the anomaly monitoring status index Dra of each production area within the current evaluation time window to the management personnel; Dra combines precision (the proportion of correct judgments) and recall (the ability to detect actual abnormal types), and forms a unified index through weighted calculation, which can comprehensively reflect the performance of the anomaly monitoring system in terms of "judgment correctness" and "missed detection risk". For example, a high precision indicates that the system has strong reliability in judging known anomalies, while a high recall indicates that the system can effectively cover the actual abnormal types that occur and avoid missed judgments; Through horizontal comparison of Dra in each production area, areas with low anomaly judgment accuracy or insufficient recall of specific abnormal types (such as low recall of "equipment overload" in a certain area) can be identified, and then the monitoring parameters (such as power threshold, time window) can be adjusted accordingly or the model can be optimized (such as updating the weight of historical operation and maintenance data), so as to solve the problems of "false judgment" or "missed judgment", providing data support for locating weak links and targeted optimization;

[0020] Embodiment 2 Please refer to Figure 2 As shown, based on an abnormal monitoring method for working condition power consumption information provided in Embodiment 1 of the present application, Embodiment 2 of the present application proposes an abnormal monitoring system for working condition power consumption information. Embodiment 2 is only a preferred manner of Embodiment 1, and the implementation of Embodiment 2 will not affect the independent implementation of Embodiment 1.

[0021] Specifically, the difference of an abnormal monitoring system for working condition power consumption information provided in Embodiment 2 of the present application is that it includes a power consumption information collection module, a power consumption model update module, an anomaly evaluation module, an anomaly type judgment module, and an anomaly result transmission module; The power consumption information collection module is used to collect the power consumption data of different production areas and the power of each power consumption equipment in different production areas in real time; The power consumption model update module is used to preset the update time interval of the normal power consumption range. After the update time interval is reached, the change range of the power consumption data of each production area under different working conditions and the change range of the power of each power consumption equipment are tested as the normal power consumption range and the normal power range, and are input into the pre-constructed power consumption model for update; The anomaly evaluation module is used to take the power consumption data of different production areas and the power data of each power consumption equipment in different production areas as each group of data points and input them into the power consumption model to compare with the corresponding normal power consumption range or normal power range. If a certain group of data points is higher than the highest value within the corresponding normal power consumption range or normal power range, it is marked as an abnormal data point; The anomaly type judgment module is used to, when abnormal data points are detected at a certain moment, after identifying the production area where the abnormal data points are located, combine historical data to execute corresponding steps to preliminarily judge the anomaly type of the abnormal data points; The abnormal result transmission module is used to trigger a warning signal and send it to the management personnel after judging the type of abnormality; The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to the specific implementation manners. Obviously, according to the content of this specification, many modifications and variations can be made. These embodiments are selected and specifically described in this specification in order to better explain the principle and practical application of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A method for abnormal monitoring of power consumption information under working conditions, characterized in that, Including: Establishing an electricity consumption model: preset the update time interval of the normal electricity consumption range. After reaching the update time interval, test the change range of the electricity consumption data of each production area under different working conditions and the change range of the power of each electrical equipment as the normal electricity consumption range and the normal power range, and input them into the pre-constructed electricity consumption model for update; Real-time data comparison: For the electricity consumption data of different production areas and the power data of each electrical equipment in different production areas, input them into the electricity consumption model as each group of data points for comparison with the corresponding normal electricity consumption range or normal power range. If a group of data points is higher than the highest value within the corresponding normal electricity consumption range or normal power range, it is marked as an abnormal data point.

2. Abnormal type judgment: When abnormal data points are detected at a certain moment, after identifying the production area where the abnormal data points are located, perform corresponding steps in combination with historical data to preliminarily judge the abnormal type of the abnormal data points.

3. The abnormal monitoring method for power consumption information under a working condition according to claim 1, wherein After identifying the production area where the abnormal data points are located, perform corresponding steps in combination with historical data, specifically: : The abnormal data points corresponding to the electricity consumption data and power are represented by the numbers a and b respectively; : If there is only one set of abnormal data points in the production area at the current moment, after identifying the number of the abnormal data points, the corresponding step or ; : If there are abnormal data points a and b in the production area at the current moment, first count the number of abnormal data points b in the production area at the current moment. If the number is one, it is determined as a single electrical equipment abnormality. If the number is greater than one, it is determined as an abnormality of multiple electrical equipment; For a single electrical equipment anomaly and a combination of anomalies of multiple electrical equipment and anomaly data a, the corresponding steps are executed or .

4. The abnormal monitoring method for working condition power consumption information according to claim 2, characterized in that Execution steps Preliminarily determine the anomaly type of the anomaly data point, specifically: : If there is only abnormal data point a in the production area at the current moment, match the current moment with the working time range of the corresponding production area. If the match fails, initially determine that the abnormal type is illegal electricity consumption; otherwise, extract the abnormal reasons for the last x times (where x > 5) that only abnormal data point a exists within the working time range from the historical data of the corresponding production area; use the abnormal reason with the highest occurrence frequency as the initially determined abnormal type.

5. The abnormal monitoring method for power consumption information under a working condition according to claim 3, characterized in that Execution steps Preliminarily determine the anomaly type of the anomaly data point, specifically: : If there is only an abnormal data point b in the production area at the current moment, further identify the electrical equipment corresponding to the abnormal data point b, and then extract the highest value within the normal power range of the corresponding electrical equipment from the power consumption model; Use the highest value within the normal power range as the power threshold for the corresponding electrical equipment, and observe the change of the abnormal data b of the corresponding electrical equipment within the set time window after the current moment; If the continuous duration of the abnormal data b of the corresponding user equipment being higher than the power threshold within the set time window exceeds the preset reference duration, preliminarily judge that the abnormal type is electrical equipment overload; If the continuous duration of the abnormal data b of the corresponding user equipment being higher than the power threshold within the set time window is lower than the preset reference duration, extract the power at each time point within the set time window according to the pre-divided time interval and substitute it into the standard deviation formula for calculation to obtain the fluctuation value of the abnormal data b of the corresponding electrical equipment within the set time window; compare the abnormal data b fluctuation value with the preset fluctuation threshold. If the abnormal data b fluctuation value is higher than the fluctuation threshold, it is determined that there is a fault inside the equipment, and at the same time, retrieve the historical operation and maintenance records of the corresponding electrical equipment; the historical operation and maintenance records include the change situation of the abnormal data point b of the corresponding electrical equipment in each historical time, the maintenance time, the maintenance personnel, and the abnormal reason, and locate the abnormal type of the corresponding electrical equipment at the current moment based on the historical operation and maintenance records.

6. The abnormal monitoring method for power consumption information under a working condition according to claim 4, characterized in that, Locating the abnormal type of the corresponding electrical equipment at the current moment based on the historical operation and maintenance records, specifically: Extract the power at each time point within the set time window according to the pre-divided time intervals, and construct the current power sequence, denoted as ; is the power at each time point within the set time window, and n is the number of sampling points within the set time window; Screen the changes of the historical abnormal data b of the corresponding electrical equipment in the historical operation and maintenance records, retain the records where the fluctuation value of the abnormal data b of the corresponding electrical equipment is higher than the fluctuation threshold, and extract the power sequence of each abnormal data b within the set time window from the retained records, expressed as ; where k is the number of each group of retained records, is the power value at the m-th time point in the k-th group of retained records; For the current power sequence Y and the power sequences of each historical group retained for recording calculate the similarity values between the power sequences of each historical group retained for recording and the current power sequence Y using cosine similarity; Obtain the current power sequence Y and the power sequences of the historical retained records for each group After obtaining the similarity values, generate a sorted list of the similarity values for each group in descending order; Select the top three historical records with the highest similarity, extract their abnormal reasons and historical occurrence frequencies, and the historical occurrence frequency is the occurrence frequency of the corresponding abnormal reason in the historical operation and maintenance records; combine the similarity weights and historical occurrence frequencies to calculate the confidence levels of the three groups of abnormal reasons. After normalizing the similarity and historical occurrence frequencies, multiply the historical occurrence frequency and similarity by the corresponding set weights respectively, and divide the result of multiplying the historical occurrence frequency by the corresponding weight by the result of multiplying the similarity by the corresponding weight to obtain the confidence levels of the three groups of abnormal reasons; Select the abnormal reason with the highest confidence level from the confidence levels of the three groups of abnormal reasons as the preliminarily judged abnormal type of the corresponding electrical equipment.

7. A method for abnormal monitoring of power consumption information under a working condition according to claim 5, characterized in that, Execution steps Preliminarily determine the type of abnormality of the abnormal data point, specifically: : For the abnormality of a single electrical equipment and the abnormal data point a, after using the steps to identify the abnormal type of the corresponding electrical equipment, it is preliminarily determined that the abnormal type is the abnormal power consumption data caused by the abnormal type of the corresponding electrical equipment.

8. A method for abnormal monitoring of power consumption information under a working condition according to claim 6, characterized in that, Execution steps Preliminarily determine the anomaly type of the anomaly data point, specifically: : For the anomalies of multiple electrical equipment and the anomaly data point a, first check the input voltage of the distribution box in the production area at the current moment. If the measured input voltage is higher than the rated voltage, initially judge that the anomaly type is a common system anomaly; otherwise, go to step After identifying the anomaly types of each abnormal electrical equipment, combine the fault data sets and initially judge that the anomaly type is the abnormal power consumption data caused by the fault data sets.

9. The abnormal monitoring method for power consumption information under a working condition according to claim 7, characterized in that, Also including: Multi-point metering layout: Real-time collect the electricity consumption data of different production areas and the power of each electrical equipment in different production areas; Abnormal warning processing: After judging the abnormal type, trigger a warning signal and send it to the management personnel.

10. A working condition power consumption information abnormal monitoring system is applied to the working condition power consumption information abnormal monitoring method proposed in any one of the above claims 1-8, and is characterized in that, Including: Power consumption information acquisition module: Real-time acquisition of power consumption data in different production areas and the power of each electrical equipment in different production areas; Power consumption model update module: Preset the update time interval of the normal power consumption range. After reaching the update time interval, test the change range of power consumption data and the change range of the power of each electrical equipment under different working conditions in each production area as the normal power consumption range and the normal power range, and input them into the pre-constructed power consumption model for update; Abnormal assessment module: Use the power consumption data in different production areas and the power data of each electrical equipment in different production areas as each group of data points and input them into the power consumption model for comparison with the corresponding normal power consumption range or normal power range. If a group of data points is higher than the highest value within the corresponding normal power consumption range or normal power range, it is marked as an abnormal data point; Abnormal type judgment module: When abnormal data points are detected at a certain moment, after identifying the production area where the abnormal data points are located, execute corresponding steps in combination with historical data to initially judge the abnormal type of the abnormal data points; Abnormal result transmission module: After judging the abnormal type, trigger a warning signal and send it to the management personnel.

Images