Search

CN-121996927-A - Working hour statistical method and system of engineering mechanical equipment and electronic equipment

CN121996927ACN 121996927 ACN121996927 ACN 121996927ACN-121996927-A

Abstract

The invention relates to the technical field of man-hour statistics, in particular to a man-hour statistics method, a system and electronic equipment of engineering machinery equipment, wherein the method comprises the steps of taking ACC switching signals as trigger conditions, starting statistics only when the signals are started, synchronously collecting multi-sensor data, extracting sliding window statistical characteristics, outputting state results and confidence coefficient through a trained decision model, checking and eliminating interference through an interference feature library, processing state conflict and undetermined state by combining erroneous judgment and eliminating rules, accumulating continuous working time exceeding a dedicated man-hour threshold as effective man-hour, and regularly adjusting and optimizing model parameters and updating an interference feature library. The invention accurately distinguishes state results through multi-sensor data fusion and intelligent judgment, improves working hour statistics accuracy, avoids environmental interference and artificial counterfeiting through interference feature library verification and misjudgment elimination, and provides reliable data for construction accounting through self-adaptive updating adaptation of multiple types of equipment and complex working conditions.

Inventors

  • CAO YILIN
  • WANG XIUGANG

Assignees

  • 苏州晟腾智联科技有限公司

Dates

Publication Date
20260508
Application Date
20260114

Claims (9)

  1. 1. A method for counting man-hours of engineering machinery equipment, comprising: Collecting ACC switch signals of engineering mechanical equipment, and executing the following steps if the ACC switch signals are in an on state: Synchronously acquiring and processing the multi-sensor data according to the sampling frequency, and extracting statistical features in the sliding window from the processed multi-sensor data according to a preset sliding window; Inputting the statistical features into a trained decision model, and outputting state results and confidence degrees of all sliding windows, wherein the state results comprise equipment states and undetermined states, and the equipment states comprise working states and non-working states; Checking the state result of each sliding window based on the interference feature library, if the checking has interference, judging that the sliding window is in a non-working state, and adjusting the confidence coefficient of the sliding window to be a preset high confidence coefficient threshold value; Based on the state result of the sliding window, marking the state result of each time point, processing the state result conflict and the undetermined state of each time point according to the misjudgment and rejection rule, accumulating the time points continuously marked as working states, and counting effective working hours when the accumulated time length is larger than the dedicated working hour threshold of the engineering mechanical equipment; And (3) periodically evaluating the judging precision of the state result, performing parameter tuning on the decision model according to feedback, and updating an interference feature library.
  2. 2. The working hour counting method of engineering machinery equipment according to claim 1, wherein the method for checking the state result of each sliding window based on the interference feature library comprises the following steps: And calling a preset interference feature library, comparing the statistical features extracted by the sliding window with the interference features in the interference feature library, judging that the sliding window has interference if the statistical features of the sliding window match the interference features, and judging that the sliding window has no interference if the statistical features of the sliding window do not match the interference features.
  3. 3. The working hour statistical method of engineering mechanical equipment according to claim 2, wherein the interference feature library is pre-stored with interference features of artificial counterfeiting and external environment interference, and the interference features are consistent with the dimension of the statistical features extracted by the sliding window.
  4. 4. A method for counting man-hours of engineering machinery according to claim 3, wherein the method for comparing the statistics extracted by the sliding window with the interference features in the interference feature library is as follows: Carrying out normalization processing on the statistical features and carrying out similarity calculation with the interference features, wherein the similarity calculation adopts cosine similarity or Euclidean distance algorithm; And if the calculated similarity value is greater than or equal to a preset similarity threshold value or the Euclidean distance value is less than or equal to a preset distance threshold value, judging that the statistical characteristic of the sliding window matches the interference characteristic.
  5. 5. The working hour counting method of engineering machinery equipment according to claim 1, wherein the method for processing the state result conflict and the pending state of each time point according to the misjudgment rejection rule comprises the following steps: summarizing all the state results covering the time point and the corresponding confidence coefficient for each time point, and determining the state result of the time point by adopting a confidence coefficient voting method; if continuous time points are marked as undetermined states, and the length of the corresponding time period is greater than or equal to a preset undetermined duration threshold value, uniformly correcting the equipment states of all the time points in the time period into non-working states; if a time point marked as a pending state exists, wherein the length of a single or continuous time period is smaller than a threshold value of the pending time period, judging that the state of the equipment belongs to according to the previous and subsequent time periods: When the device states of the time points or the continuous time periods are different, the device states of the time points in the time points or the continuous time periods and the corresponding confidence levels are summarized, and the device states of the time points in the undetermined state or the undetermined state time points in the continuous time periods are determined by adopting a confidence voting method.
  6. 6. The method of claim 5, wherein the confidence voting method includes accumulating the confidence levels corresponding to the respective state results for each time point, obtaining a sum of the confidence levels corresponding to the respective state results, and marking the time point with the state result having the highest sum of the confidence levels.
  7. 7. The working hour counting method of engineering machinery equipment according to claim 1, wherein the dedicated working hour threshold of the engineering machinery equipment is determined based on equipment type, operation characteristics and preset working condition requirements and is used for providing a judging standard of effective working hours.
  8. 8. A man-hour statistical system of engineering machinery equipment, comprising: The ACC signal prior module is used for judging the running state of the equipment according to the ACC switching signal so as to determine start-stop logic of subsequent working hour statistics; The acquisition processing module is used for synchronously acquiring and processing the multi-sensor data according to the sampling frequency and extracting statistical characteristics in the sliding window from the processed multi-sensor data according to a preset sliding window; the decision module is used for inputting the statistical characteristics into a trained decision model and outputting the state results and the confidence coefficient of each sliding window, wherein the state results comprise equipment states and undetermined states, and the equipment states comprise working states and non-working states; The interference verification module is used for verifying the state result of each sliding window based on the interference feature library, judging that the sliding window is in a non-working state if the interference exists in the verification, and not changing the state result of the sliding window if the interference does not exist; The working hour counting module is used for marking the state result of each time point based on the state result of the sliding window, processing the state result conflict and the undetermined state of each time point according to the misjudgment and rejection rule, accumulating the time points continuously marked as working states, and counting effective working hours when the accumulated time length is larger than the exclusive working hour threshold value of the engineering mechanical equipment; And the self-adaptive updating module is used for periodically evaluating the judging precision of the state result, carrying out parameter tuning on the decision model according to feedback and updating the interference feature library.
  9. 9. An electronic device, characterized in that the electronic device comprises: One or more processors; a memory for storing one or more programs; when executed by the one or more processors, causes the one or more processors to implement the method of man-hour statistics of a work machine according to any one of claims 1-7.

Description

Working hour statistical method and system of engineering mechanical equipment and electronic equipment Technical Field The present invention relates to the field of man-hour statistics technologies, and in particular, to a man-hour statistics method and system for engineering machinery equipment, and an electronic device. Background Engineering mechanical equipment is widely applied to the engineering such as earth and stone sides, road construction and the like, and the actual operation man-hour directly influences the construction cost and efficiency. The traditional working hour statistics usually depend on manual recording or single sensor monitoring, the problems of neglected recording, false alarm and the like are easy to occur in manual registration, and the early mechanical timer can only monitor single parameters such as the rotating speed of an engine, has low monitoring precision and poor adaptability and can not effectively solve the working hour timing problem. At present, some Beidou equipment, GPS positioners and other equipment judge whether an engine is started or not to count the working time by monitoring ACC signals, but the ACC can only reflect the state of the engine, can not distinguish idle idling from actual operation, and can not resist surrounding vibration interference or artificial intervention. On the other hand, under the complex working condition, misjudgment or missed judgment is easy to generate only depending on single vibration, speed or inclination angle data. The existing research shows that the multi-sensor fusion can remarkably improve the reliability and accuracy of the detection result, but the existing engineering machinery man-hour monitoring system mostly adopts simple threshold judgment, lacks intelligent analysis and self-adaptation capability, and is difficult to meet the requirements of eliminating interference and preventing counterfeiting. Therefore, there is a need for a working hour statistical method that can improve working hour monitoring accuracy, eliminate interference, and have adaptive adjustment capability for different types of engineering machinery. Disclosure of Invention The invention aims to provide a working hour statistical method, a working hour statistical system and electronic equipment of engineering machinery equipment, and aims to solve the problems of inaccurate equipment state identification, weak anti-interference capability and poor equipment suitability in the prior art. The technical scheme of the invention is that the working hour statistical method of engineering mechanical equipment comprises the following steps: Collecting ACC switch signals of engineering mechanical equipment, and executing the following steps if the ACC switch signals are in an on state: Synchronously acquiring and processing the multi-sensor data according to the sampling frequency, and extracting statistical features in the sliding window from the processed multi-sensor data according to a preset sliding window; Inputting the statistical features into a trained decision model, and outputting state results and confidence degrees of all sliding windows, wherein the state results comprise equipment states and undetermined states, and the equipment states comprise working states and non-working states; Checking the state result of each sliding window based on the interference feature library, if the checking has interference, judging that the sliding window is in a non-working state, and adjusting the confidence coefficient of the sliding window to be a preset high confidence coefficient threshold value; Based on the state result of the sliding window, marking the state result of each time point, processing the state result conflict and the undetermined state of each time point according to the misjudgment and rejection rule, accumulating the time points continuously marked as working states, and counting effective working hours when the accumulated time length is larger than the dedicated working hour threshold of the engineering mechanical equipment; And (3) periodically evaluating the judging precision of the state result, performing parameter tuning on the decision model according to feedback, and updating an interference feature library. Preferably, the method for checking the status result of each sliding window based on the interference feature library comprises the following steps: And calling a preset interference feature library, comparing the statistical features extracted by the sliding window with the interference features in the interference feature library, judging that the sliding window has interference if the statistical features of the sliding window match the interference features, and judging that the sliding window has no interference if the statistical features of the sliding window do not match the interference features. Preferably, the interference feature library is pre-stored with interference features of artificial counterfeiting