CN-122016875-A - Method and system for determining metal content of ore based on knowledge graph

CN122016875ACN 122016875 ACN122016875 ACN 122016875ACN-122016875-A

Abstract

The invention discloses a method and a system for measuring the metal content of ores based on a knowledge graph, and relates to the technical field of intelligent detection, and the method comprises the following steps of establishing an energy spectrum acquisition baseline, continuously acquiring signals generated by detection equipment in a high-energy density area in the metal content measuring process, recording tailing morphology and peak foot extension change, and generating an energy spectrum tailing data band; and extracting the starting time, the ending time and the position with the maximum energy change of each energy spectrum peak based on the energy spectrum tailing data band, and generating a characteristic point set. The method realizes signal dynamic coordination by establishing the energy spectrum acquisition base line and the time control chain, inhibits peak shape expansion caused by tailing effect, ensures the true and stable of the energy spectrum peak shape, identifies and eliminates virtual false peaks through the characteristic point set, generates a real peak corresponding relation list, improves the structural consistency of energy spectrum data and the accuracy of knowledge graph reasoning, and realizes the accurate determination of metal content.

Inventors

NIU ZHIJIAN
WANG YE
CHEN ZHENGHUI

Assignees

中国地质科学院

Dates

Publication Date: 20260512
Application Date: 20260112

Claims (10)

1. The method for determining the metal content of the ore based on the knowledge graph is characterized by comprising the following steps of: Establishing an energy spectrum acquisition baseline, continuously acquiring signals generated by detection equipment in a high-energy density region in the metal content measurement process, recording tailing morphology and peak foot extension change, and generating an energy spectrum tailing data band; extracting the starting time, the ending time and the position with the maximum energy change of each energy spectrum peak based on the energy spectrum tailing data band, and generating a characteristic point set; according to the characteristic point set, carrying out point-to-point comparison on the ascending section and the descending section of each energy spectrum peak, separating and removing false peaks caused by tailing effect, and generating a real metal energy spectrum peak corresponding relation list; Establishing a time control chain for signal acquisition and data matching based on a real metal energy spectrum peak corresponding relation list, determining a signal exposure rhythm and a reading interval according to the time control chain, and generating a time control instruction for inhibiting tailing interference; and executing short-time static, delay compensation and reverse energy release operation in a signal attenuation stage according to the time control instruction, and carrying out dynamic energy adjustment on the energy spectrum signal so as to weaken and keep controlled a tailing effect in an energy transmission process.
2. The method for determining the metal content of the ore based on the knowledge graph according to claim 1, wherein the step of generating the energy spectrum tailing data band is as follows: aiming at the energy response characteristic of the detection equipment in the metal content measurement process, determining the energy coverage range and sampling time resolution of signal acquisition, and continuously acquiring a complete signal response curve in a high energy density region to form an energy spectrum response sequence; Dynamically tracking and continuously recording the trailing form of the energy spectrum response sequence in a high-energy section, dividing an energy change interval into a plurality of time windows, and recording an energy value, a change rate and an attenuation duration; uniformly rearranging energy change information in a time window according to the energy attenuation direction to generate a trailing extension mapping sequence corresponding to an energy response curve of the detection equipment; And integrating the continuously recorded data according to the time sequence and the energy distribution rule of signal acquisition to form an energy spectrum tailing data band.
3. The method for determining metal content of ore based on knowledge-graph according to claim 2, wherein the feature point set generating step comprises the following steps: After forming an energy spectrum tailing data band, carrying out layering analysis on continuous energy response information contained in the energy spectrum tailing data band according to a time sequence, and generating an energy distribution sequence with equal time intervals so as to establish a peak start-stop time boundary frame; in a time sequence structure of energy distribution, identifying an energy inflection point of a peak-shaped rising section according to the change trend of energy along with time, and determining the moment when the energy response starts to be obviously increased as the starting time of an energy spectrum peak; tracking an energy attenuation process, identifying the end time of an energy spectrum peak, and determining a time sequence corresponding structure of the peak shape by comparing the continuous relation of a peak shape ascending section and a peak shape descending section on a time axis; after the starting time and the ending time are defined, the position with the largest energy change is identified according to the energy change track, and the starting time, the ending time and the position with the largest energy change are integrated to generate the characteristic point set.
4. The method for determining metal content of ore based on knowledge graph according to claim 3, wherein in the process of generating the feature point set, the position of the maximum energy change is limited in the energy peak top interval of the energy spectrum peak, and the energy change track between the starting time and the ending time is synchronously associated with the energy attenuation process of the peak foot extension area, so that the feature point set reflects the whole process of rising, peak top and attenuation of peak shape energy at the same time.
5. The method for determining metal content of ore based on knowledge graph according to claim 3, wherein the step of generating the real metal energy spectrum peak correspondence list is as follows: The time boundary information of each energy spectrum peak is segmented and arranged according to the characteristic point set, the energy spectrum peak is divided into energy sequences according to key positions within the range from the starting time to the ending time, and a continuous corresponding relation is formed between the ascending section and the descending section on the energy change trend; The ascending section and the descending section are compared point by point, trailing characteristics of energy expansion of energy spectrum peaks are identified according to energy change trend, and continuity of energy change and breakpoint positions are judged to distinguish real peaks from trailing peaks; Separating an energy interval with tailing effect, removing false peaks, and removing a tailing part from an independent peak set through the difference of the energy change direction and the energy amplitude; and correlating the reserved position with the maximum starting time, ending time and energy variation of the real peak to form a real metal energy spectrum peak corresponding relation list.
6. The method for determining metal content of ore based on a knowledge graph according to claim 5, wherein when the ascending section and the descending section are compared point by point, the energy change direction and the time continuity are used as comparison basis, the energy residual range of the tailing region is determined by identifying the local energy re-ascending phenomenon in the energy attenuation region, and when the false peak is eliminated, the extension signal in the tailing region is separated from the energy spectrum data according to the overlapping relation of the start time and the stop time of the adjacent peak.
7. The method for determining metal content of ore based on a knowledge graph according to claim 5, wherein the time control instruction for suppressing tailing interference is generated by: The time information of each real peak is concentrated and tidied according to the real metal energy spectrum peak corresponding relation list, and the position with the maximum change of the starting time, the ending time and the energy is used as a time reference point to form a time sequence frame; constructing a time control chain of a signal acquisition stage according to a time sequence frame, arranging start-stop intervals of each real peak according to energy response intensity and duration, and setting a buffer time interval between adjacent peaks; determining a signal exposure rhythm and a reading interval according to the time distribution relation and the energy response rule of each node in the time control chain, so that the signal acquisition process and the energy change trend of the energy spectrum peak are kept synchronous; generating a time control instruction according to the overall structure of the time control chain, and integrating the acquisition action, the energy response state and the reading time sequence into a continuous time instruction sequence.
8. The method for determining metal content of ore based on a knowledge graph according to claim 7, wherein the generation of the time control instruction includes arranging the signal acquisition start instruction, the exposure continuation instruction, the signal reading instruction and the acquisition end instruction in order according to the time nodes in the real metal spectrum peak correspondence list so that each instruction corresponds to a position where the start time, the end time and the energy of the real peak are most varied.
9. The knowledge-graph-based ore metal content measuring method according to claim 7, wherein the steps of performing short-time rest, delay compensation and reverse energy release operations in sequence in a signal attenuation stage according to a time control instruction, and performing dynamic energy adjustment on the energy spectrum signal are as follows: Under the guidance of a time control instruction, carrying out initial static control on the energy response process of the signal attenuation stage, so that the signal receiving state is switched to a short-time static state to release the internal accumulated charge; After the short-time rest operation is finished, executing delay compensation operation, and adjusting a signal acquisition interval according to a time control instruction so as to keep the energy release rhythm and the attenuation speed synchronous; performing a reverse energy release operation after the delay compensation operation, counteracting the retained energy by a reverse energy release pulse to accelerate the energy release process; and continuously linking short-time rest and delay compensation with reverse energy release operation according to the time control instruction, and carrying out dynamic energy adjustment on the energy spectrum signal so as to realize the controlled balance of the energy transmission process.
10. The knowledge-graph-based ore metal content measuring system for realizing the knowledge-graph-based ore metal content measuring method according to any one of claims 1 to 9, characterized by comprising an energy spectrum acquisition baseline construction module, a characteristic point extraction module, a false peak identification and elimination module, a time control chain generation module and a dynamic energy adjustment module: The energy spectrum acquisition baseline construction module is used for establishing an energy spectrum acquisition baseline, continuously acquiring signals generated by detection equipment in a high energy density region in the metal content measurement process, recording tailing morphology and peak foot extension change, and generating an energy spectrum tailing data band; the characteristic point extraction module is used for extracting the starting time, the ending time and the position with the maximum energy change of each energy spectrum peak based on the energy spectrum trailing data band to generate a characteristic point set; The false peak identification and elimination module is used for carrying out point-to-point comparison on the ascending section and the descending section of each energy spectrum peak according to the characteristic point set, separating and eliminating false peaks caused by tailing effect, and generating a real metal energy spectrum peak corresponding relation list; The time control chain generation module establishes a time control chain for signal acquisition and data matching based on a real metal energy spectrum peak corresponding relation list, determines a signal exposure rhythm and a reading interval according to the time control chain, and generates a time control instruction for inhibiting trailing interference; And the dynamic energy adjusting module is used for executing short-time static, delay compensation and reverse energy release operation in a signal attenuation stage according to the time control instruction, and carrying out dynamic energy adjustment on the energy spectrum signal so as to weaken and keep controlled a tailing effect in the energy transmission process.

Description

Method and system for determining metal content of ore based on knowledge graph Technical Field The invention relates to the technical field of intelligent detection, in particular to a method and a system for determining metal content of ores based on a knowledge graph. Background The determination of the metal content of the ore based on the knowledge graph refers to semantic association and structural modeling of multi-source ore data such as traditional chemical analysis, spectrum detection, microelement analysis, geological exploration and the like through the knowledge graph technology, so that intelligent reasoning and accurate identification between the ore components and the metal content are realized on the knowledge level. The method comprises the steps of firstly constructing a multidimensional knowledge node system comprising mineral types, element characteristics, an ore forming environment, detection parameters and experimental results, then establishing a causal relationship network according to association weights among nodes, automatically identifying and correcting detection signals of unknown ore samples through a semantic reasoning and characteristic matching algorithm, and finally generating quantitative determination results of metal content under the guidance of a knowledge graph. Compared with the traditional measurement mode based on single detection, the method can realize self-adaptive analysis and high-precision content assessment of complex ore samples by utilizing a reasoning and tracing mechanism between knowledge layers. In the metal energy spectrum analysis process of the prior art, when the detection equipment performs signal acquisition in a high-energy density area, the tailing effect caused by response lag of a detector, signal accumulation unbalance or charge transfer residues is easy to occur, so that the energy spectrum peak originally corresponding to a single metal element is stretched and expanded in a high-energy section to form a pseudo-continuous peak column. Because the knowledge graph node matching logic in the prior art generally relies on a fixed threshold value and a static characteristic template to perform peak identification, when the tailing signals are encountered, the system is difficult to distinguish the energy difference between a real peak and a tailing peak, and the tailing peak of the same element is easy to be misjudged as an independent metal characteristic node, so that an incorrect element attribution relation is formed in a graph structure. The problem not only can cause the virtual increase of metal types, but also can destroy the semantic consistency of energy spectrum data, so that a knowledge reasoning chain is deviated, and the accuracy and the reliability of metal content measurement are seriously affected. The above information disclosed in the background section is only for enhancement of understanding of the background of the disclosure and therefore it may include information that does not form the prior art that is already known to a person of ordinary skill in the art. Disclosure of Invention The invention aims to provide a method and a system for determining metal content of ores based on a knowledge graph, so as to solve the problems in the background technology. In order to achieve the purpose, the invention provides the following technical scheme that the method for determining the metal content of the ore based on the knowledge graph comprises the following steps: Establishing an energy spectrum acquisition baseline, continuously acquiring signals generated by detection equipment in a high-energy density region in the metal content measurement process, recording tailing morphology and peak foot extension change, and generating an energy spectrum tailing data band for feature identification; extracting the starting time, the ending time and the position with the maximum energy change of each energy spectrum peak based on the energy spectrum tailing data band, and generating a characteristic point set for determining the tailing range; according to the characteristic point set, carrying out point-to-point comparison on the ascending section and the descending section of each energy spectrum peak, separating and removing false peaks caused by tailing effect, and generating a real metal energy spectrum peak corresponding relation list; Establishing a time control chain for signal acquisition and data matching based on a real metal energy spectrum peak corresponding relation list, determining a signal exposure rhythm and a reading interval according to the time control chain, and generating a time control instruction for inhibiting tailing interference; And executing short-time rest, delay compensation and reverse energy release operation in a signal attenuation stage according to the time control instruction, and carrying out dynamic energy adjustment on the energy spectrum signal, so that the tailing effect is weakened and ke