Monitoring and alarming method, monitoring and alarming device, and computer-readable storage medium

Abstract

The invention discloses a monitoring alarming method. The monitoring alarming method comprises the steps of receiving a data stream including historical data associated with target parameters to be monitored; utilizing a dynamic linear model to analyze the data stream to obtain prediction parameters associated with the target parameters in a later first time period; counting the target parametersin the first time period to obtain actual parameters associated with the target parameters in the first time period; comparing the prediction parameters with the actual parameters to judge whether ornot an alarming signal is generated. The invention further discloses a corresponding computer-readable storage medium and a corresponding monitoring alarming device. According to the scheme, monitoring alarming is conducted by integrating long-term accumulated historical monitoring data and utilizing a prediction model generated on the basis of machine learning, a reasonable interval is set on thebasis of the historical data without only relying on simple experience judgment any more, and manual intervention, false alarm situations and operation and maintenance costs are greatly reduced.

Description

technical field [0001] The disclosure belongs to the field of monitoring and alarming, and in particular relates to a monitoring and alarming method, a monitoring and alarming device and a corresponding tangible computer-readable storage medium. Background technique [0002] Traditional monitoring systems are mainly based on rules for monitoring and alarming, and many of the rules used here are formulated based on the experience of operation and maintenance personnel, and during the use of these rules, they need to be continuously updated in the subsequent operation and maintenance process Manually intervene and modify thresholds to accommodate real-time monitoring requirements. At the same time, the general setting of these rules is too single and rigid, which makes misreporting very common, which will inevitably lead to a sharp increase in operation and maintenance costs. [0003] figure 1 A flow chart of a method for monitoring alarms in the prior art is shown, and it c...

AI Technical Summary

Problems solved by technology

At the same time, the general setting of these rules is too single and rigid, which makes misreporting very common, which will inevitably lead to a sharp increase in operation and maintenance costs

[0003] figure 1 A flow chart of a method for monitoring alarms in the prior art is shown, and it can be seen from the flow chart that the monitoring alarm method 100 only includes two steps, that is, the data flow to be monitored is first received in method step 110 As well as the judgment rules entered manually, the judgment rules here are manually entered, that is to say, the judgment rules here are set entirely based on the experience of the operation and maintenance personnel, and the operation and maintenance personnel will determine the judgment rules when the business fluctuates. Make adaptive modifications, but such modifications are only empirical rather than adjusted based on objective real-time data, so such adjustments are very random, and such modifications may not necessarily meet the monitoring and alarm requirements. Requirement; Next, in method step 120, the received data flow is judged according to the input judgment rule, to judge whether to output the alarm signal

Since the judgment rules are pre-set, such monitoring alarms will inevitably lead to a very common situation of false alarms, which in turn will inevitably lead to a sharp increase in operation and maintenance costs

Images