Search

CN-122022661-A - Pulp paper cloud warehouse management system based on Internet of things

CN122022661ACN 122022661 ACN122022661 ACN 122022661ACN-122022661-A

Abstract

The invention relates to the technical field of storage management of the Internet of things, and discloses a pulp paper cloud warehouse management system based on the Internet of things. The system comprises a data acquisition unit, a data separation unit, a mode discovery unit, a track analysis unit, a collaborative analysis engine, an anomaly detection unit and a decision logic unit. The system separates an environment parameter from a logistics information substream by collecting an environment monitoring data stream, a pattern discovery module analyzes the periodic fluctuation of the environment parameter to extract a trend pattern, a track analysis module identifies a logistics path key point extraction path pattern, a collaborative analysis engine carries out association calculation on the two types of patterns to generate a fusion pattern description, an abnormality detection module diagnoses the difference degree of the fusion pattern and a reference pattern library, and a decision logic unit locates a root cause and generates a control instruction according to the difference degree. The scheme realizes the deep association perception of the environment and the logistics state, can discover the composite risk early and automatically generate the accurate disposal strategy, and improves the intelligent level of warehouse management and the risk handling capability.

Inventors

  • SU WEI
  • SHI QIANG
  • Lin Gangxiao

Assignees

  • 厦门佰京数字科技有限公司

Dates

Publication Date
20260512
Application Date
20251204

Claims (10)

  1. 1. Pulp paper cloud warehouse management system based on thing networking, its characterized in that, the system includes: the data acquisition module is used for continuously acquiring standardized monitoring data flow of the pulp paper storage environment; the data separation module is used for separating an environment parameter substream and a logistics information substream from the standardized monitoring data stream; The mode discovery module is used for calling a mode discovery routine to perform periodic fluctuation analysis on the environmental parameter substream and extracting an environmental trend mode; the track analysis module is used for running in parallel with the mode discovery module, calling a track analysis routine to identify the path key points of the logistics information substreams, and extracting a logistics path mode; the collaborative analysis engine is used for inputting the environmental trend mode and the logistics path mode into the collaborative analysis engine to perform mode relevance calculation and generate a fusion mode description; the anomaly detection module is used for starting an anomaly detection program based on the difference degree between the fusion mode description and the reference mode library; And the decision logic unit is used for determining the root cause position according to the output of the abnormality detection program through the decision logic unit and generating a control instruction set.
  2. 2. The internet of things-based pulp and paper cloud warehouse management system of claim 1, wherein the call pattern discovery routine performs periodic fluctuation analysis on the environmental parameter substream comprising: dividing the environmental parameter substream into time slices with fixed duration; trend fitting is carried out on the temperature readings and the humidity readings in each time segment, so that a local trend line is obtained; the local trend lines of all time segments are aggregated, and a global environment trend map is constructed; identifying recurring wave cycle and amplitude features from the global environmental trend profile; The period and amplitude characteristics of the fluctuations are encoded as the environmental trend pattern.
  3. 3. The internet of things-based pulp and paper cloud warehouse management system of claim 1, wherein the invoking a trajectory resolution routine to perform path keypoint identification on the logistics information substream comprises: Analyzing a material displacement signal sequence in the logistics information substream, and restoring a material movement track; dividing the material moving track into a plurality of path sections, and calculating the average speed and direction change of each path section; detecting connection points between path segments, and marking the connection points as key nodes; Extracting the stay time and steering angle information of each key node; And generating the logistics path mode based on the key node and the attribute thereof.
  4. 4. The internet of things-based pulp and paper cloud warehouse management system of claim 1, wherein the inputting the environmental trend pattern and the logistic path pattern into a collaborative analysis engine for pattern relevance computation comprises: assigning an environmental pattern identifier to the environmental trend pattern and assigning a logistic pattern identifier to the logistic path pattern; constructing a time alignment mapping table of the environment mode and the logistics mode; calculating the co-occurrence probability of the environment mode identifier and the logistics mode identifier in the same time interval; Generating a pattern association weight matrix based on the co-occurrence probability; And carrying out weighted splicing on the environment trend mode and the logistics path mode by using the mode association weight matrix to form the fusion mode description.
  5. 5. The internet of things-based pulp and paper cloud warehouse management system of claim 1, wherein the initiating an anomaly detection procedure based on the degree of difference of the fusion pattern description from a benchmark pattern library comprises: Extracting a reference pattern library under normal operation from historical data, wherein the reference pattern library comprises a plurality of standard pattern templates; Calculating similarity scores of the fusion pattern descriptions and each standard pattern template; Determining whether the lowest similarity score is below a preset variance threshold; Triggering the anomaly detection procedure when the difference threshold is lower than the difference threshold; And the abnormality detection program scans the abnormal characteristic segments in the fusion mode description and outputs an abnormality marking sequence.
  6. 6. The internet of things-based pulp and paper cloud warehouse management system of claim 5, wherein the determining, by the decision logic unit, the root cause location from the output of the anomaly detection program comprises: receiving the abnormality marking sequence and analyzing the occurrence time point and type of the abnormality; querying a causal knowledge base, wherein the causal knowledge base stores causal relation chains between environmental events and logistic events; matching the abnormal mark sequence with event pairs in a causal knowledge base, and finding out potential cause events; Carrying out confidence evaluation on the potential cause event, and screening out a high confidence cause set; and mapping the high confidence factor set back to the cloud bin physical position to generate a root factor position report.
  7. 7. The internet of things-based pulp and paper cloud warehouse management system of claim 6, wherein the query causal knowledge base comprises: Maintaining a graph structure database, wherein nodes represent environmental or logistic events and edges represent causal relationships; Inputting event types in an abnormal mark sequence, and executing multi-hop traversal inquiry in the graph structure database; Collecting all reachable causal paths, and recording path weights; sorting the causal paths according to the path weights, and selecting top-k paths as candidate causal chains; and verifying the coincidence degree of the candidate causal link and the current data, and reserving the causal link passing verification.
  8. 8. The internet of things-based pulp and paper cloud warehouse management system of claim 1, wherein the generating the control instruction set comprises: Identifying an affected cloud bin area according to the root cause position report; Searching a history control record of the area, and finding out effective control actions in similar scenes; Adjusting control action parameters by combining current real-time data to generate personalized control instructions; Conflict detection is carried out on the personalized control instruction, so that the personalized control instruction is ensured not to conflict with the global strategy; And packaging the conflict-free control instructions into a control instruction set.
  9. 9. The internet of things-based pulp and paper cloud warehouse management system of claim 1, wherein the continuously collecting standardized monitoring data streams of a pulp and paper storage environment comprises: continuously collecting original monitoring data of a pulp storage environment through a plurality of Internet of things sensing nodes deployed in a cloud bin, wherein the original monitoring data comprise temperature readings, humidity readings and material displacement signals; Performing data reliability evaluation processing on the original monitoring data, removing abnormal data points beyond a reasonable range, and generating a checked data sequence; and inputting the checked data sequence into a data normalization module for unit unification and scaling processing, and outputting a standardized monitoring data stream.
  10. 10. The internet of things-based pulp and paper cloud warehouse management system of claim 9, wherein the performing a data reliability assessment process on the raw monitoring data comprises: Setting a data validity rule base, wherein the rule base comprises a temperature threshold range, a humidity threshold range and a signal intensity standard; comparing the original monitoring data with rules in the data validity rule base item by item, and marking data points which do not accord with the rules as suspicious data; Applying a sliding window consistency check to the suspicious data to identify isolated outliers and continuous outlier segments; removing the isolated abnormal points and the continuous abnormal segments, and adopting adjacent data smooth interpolation processing for the missing positions; and outputting the processed data sequence as the verified data sequence.

Description

Pulp paper cloud warehouse management system based on Internet of things Technical Field The invention relates to the technical field of storage management of the Internet of things, in particular to a pulp paper cloud warehouse management system based on the Internet of things. Background In the warehouse logistics of pulp paper products, the quality of the products is highly dependent on the stability of the storage and circulation environment. Abnormal fluctuation of environmental parameters such as temperature and humidity can directly influence key indexes such as moisture content and flatness of paper, and the efficiency and rationality of a logistics path are related to delivery cycle and operation cost. The existing warehouse management system generally adopts the internet of things technology to collect data, but the environment monitoring and the logistics management are mostly subsystems which independently operate. The existing technical scheme generally sets the static alarm threshold of the environmental parameter independently and carries out simple overtime monitoring on the logistics node. This monitoring approach has inherent drawbacks. The environmental monitoring system may alarm when the parameter exceeds a fixed threshold, but cannot perceive a slow deterioration trend that deviates from normal rules without overrun. The logistics system can only feed back that goods stay at a certain node, but cannot relate the stay to the dynamic change of the microenvironment. The data island phenomenon causes that the system only can present discrete abnormal phenomena, and the dynamic association between the environmental factors and the logistics operation cannot be obtained from the global view, so that effective early warning and root cause analysis are difficult to carry out. The current technical bottleneck is the lack of a mechanism capable of communicating environmental and logistic information and extracting dynamic behavior patterns from continuous data streams for correlation analysis. The system cannot send out early warning by identifying abnormal pattern association before the potential problem is evolved into substantial loss. Meanwhile, when a compound problem occurs, the root cause of the problem is caused by the failure of environmental control, the logic error of scheduling or the synergistic effect of the two, so that the treatment decision is lagged and the pertinence is insufficient. Disclosure of Invention The invention aims to provide a pulp paper cloud warehouse management system based on the Internet of things, so as to solve the problems in the background technology. In order to achieve the above object, the present invention provides a pulp paper cloud warehouse management system based on the internet of things, the system comprising: the data acquisition module is used for continuously acquiring standardized monitoring data flow of the pulp paper storage environment; the data separation module is used for separating an environment parameter substream and a logistics information substream from the standardized monitoring data stream; The mode discovery module is used for calling a mode discovery routine to perform periodic fluctuation analysis on the environmental parameter substream and extracting an environmental trend mode; the track analysis module is used for running in parallel with the mode discovery module, calling a track analysis routine to identify the path key points of the logistics information substreams, and extracting a logistics path mode; the collaborative analysis engine is used for inputting the environmental trend mode and the logistics path mode into the collaborative analysis engine to perform mode relevance calculation and generate a fusion mode description; the anomaly detection module is used for starting an anomaly detection program based on the difference degree between the fusion mode description and the reference mode library; And the decision logic unit is used for determining the root cause position according to the output of the abnormality detection program through the decision logic unit and generating a control instruction set. Preferably, the calling pattern discovery routine performs periodic fluctuation analysis on the environmental parameter substream, including: dividing the environmental parameter substream into time slices with fixed duration; trend fitting is carried out on the temperature readings and the humidity readings in each time segment, so that a local trend line is obtained; the local trend lines of all time segments are aggregated, and a global environment trend map is constructed; identifying recurring wave cycle and amplitude features from the global environmental trend profile; The period and amplitude characteristics of the fluctuations are encoded as the environmental trend pattern. Preferably, the calling the trajectory analysis routine to identify the path key point of the logistics information substream inclu