Quantitative detection method and system for components of bauxite

Abstract

The invention discloses a bauxite component quantitative detection method and system, and relates to the technical field of bauxite component quantitative detection, and the method comprises the following steps: building a dust dynamic sensing reference around a light path area of an intelligent sensing system, and extracting stable fluctuation characteristics of environment background light intensity through continuous high-frequency collection of incident light intensity along with time; and on the basis of the established dust dynamic sensing reference, synchronously comparing the reflected light intensity change acquired in real time, and when the rhythm of the reflected light intensity change deviates from a stable fluctuation characteristic, judging that the currently acquired data contains an environment shielding factor, and marking the corresponding spectral data as a dust influence state. Through trend constraint and continuous connection fusion, smooth characteristic peak change is maintained, and dynamic feedback adjustment is realized, so that the detection precision and data consistency are maintained under the dust disturbance condition, and the stability and reliability of quantitative detection of bauxite components are improved.

Description

Technical Field

[0001] This invention relates to the field of quantitative detection technology for bauxite components, specifically to methods and systems for quantitative detection of bauxite components. Background Technology

[0002] Quantitative analysis of bauxite composition refers to the process of accurately measuring and calculating the proportions of major chemical components (such as alumina, silica, iron oxide, titanium oxide, calcium oxide, and magnesium oxide) in a bauxite sample to obtain their mass or mole fraction in the overall ore. In practice, an intelligent sensing system is introduced to construct a multi-parameter synchronous acquisition chain. By fusing multiple detection methods such as spectral sensing, X-ray fluorescence, near-infrared reflectance, and pyroelectric analysis, the energy spectrum response, reflectance characteristics, and electrochemical signals of elements in micro-areas of the ore sample are captured and modeled in real time. Then, embedded algorithms are used to achieve quantitative inversion and ratio correction of the content of each element, thereby outputting high-precision composition distribution results. This method can significantly improve the automation and intelligence level of bauxite quality assessment, graded utilization, and smelting batching.

[0003] The existing technology has the following shortcomings: In the quantitative analysis of bauxite composition, when the dust concentration in the detection environment suddenly increases within a short period, fine particles suspended in the air can easily enter the optical path area of ​​the intelligent sensing system, forming a momentary light-blocking zone. At this time, the incident and reflected light signals are affected by non-uniform scattering and partial absorption, causing a sudden drop in the spectral reflectance collected by the system. Within a millisecond-level sampling period, the intelligent sensing system can easily misinterpret this optical attenuation as an abnormal change in the surface reflectance of the mineral sample, resulting in abnormal jumps in the intensity ratio of elemental characteristic peaks. This phenomenon can further lead to instability in the composition inversion algorithm, causing short-term deviations or fluctuations in the quantitative results, thus affecting the continuity and reliability of the dynamic detection chain. In batch detection scenarios, it may even cause the accumulation of overall composition determination errors, severely weakening the stability and accuracy of the detection system.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for quantitative detection of bauxite components to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for quantitative detection of bauxite components, comprising the following steps: Step 1: Establish a dynamic dust sensing benchmark around the optical path area of ​​the intelligent sensing system. By continuously and frequently collecting incident light intensity over time, extract the stable fluctuation characteristics of the ambient background light intensity and use these stable fluctuation characteristics as a reference benchmark for identifying the dust occlusion state. Step 2: Based on the established dust dynamic sensing benchmark, the real-time collected changes in reflected light intensity are synchronously compared. When the rhythm of reflected light intensity changes deviates from the stable fluctuation characteristics, it is determined that the current collected data contains environmental shading factors, and the corresponding spectral data is marked as the dust influence state. Step 3: For the spectral data that has been marked as being affected by dust, trend constraint processing is applied to the changes in the intensity of the elemental characteristic peaks within this time period. Based on the stable fluctuation characteristics, the amplitude of abnormal changes is limited, and sudden attenuation is included in the range of continuous changes, thereby suppressing the instantaneous instability of the characteristic peak intensity ratio. Step 4: Based on the characteristic peak intensity sequence after trend constraint processing, the component change rhythm of adjacent time points is continuously connected, and the data corresponding to the dust-affected section is continuously fused with the data of the unaffected section to obtain consistent component quantification results in the time dimension. Step 5: Input the obtained continuous component quantitative results back into the dust dynamic sensing benchmark update process, and dynamically adjust the judgment range of the dust dynamic sensing benchmark in combination with stable fluctuation characteristics to maintain the quantitative detection accuracy and stability of the detection system under continuous dust disturbance conditions.

[0007] Preferably, the steps for establishing a dynamic dust sensing baseline around the optical path area of ​​the intelligent sensing system include: Accurately confirm the geometric layout of the optical path region, determine the relative positional relationship between the incident light source emitting end and the reflected light receiving end, and establish a light intensity acquisition channel on the beam propagation path. Record the change of incident light intensity over time through continuous high-frequency sampling to form an initial light intensity dataset for the optical path region. Long-term high-frequency sampling of incident light intensity is performed, the sampling results are time-aligned, stable segments are selected and the maximum, minimum and average values ​​of light intensity are statistically analyzed, and the stable fluctuation characteristics of ambient background light intensity are extracted to form a continuous time light intensity dataset. A dynamic dust sensing benchmark is established based on stable fluctuation characteristics. The incident light intensity is continuously collected and the sampled values ​​are compared with the range of change of stable fluctuation characteristics point by point. When the sampled value exceeds the limit range, it is marked as a dust-affected state, forming a judgment result of the optical path environment state. Real-time incident light intensity data is incorporated into the dust dynamic sensing benchmark, and the stable fluctuation characteristics are dynamically expanded by combining the newly acquired light intensity fluctuation range, so as to maintain the effectiveness and temporal continuity of the dust dynamic sensing benchmark during the detection process.

[0008] Preferably, during the dynamic expansion of the dust dynamic sensing benchmark, real-time incident light intensity data and stable fluctuation characteristics are correlated according to time index, and the judgment range of light intensity fluctuation range is adjusted according to the change of dust concentration in the detection environment, so that the dust dynamic sensing benchmark maintains the temporal continuity and the stability of optical path state recognition under environmental disturbance conditions.

[0009] Preferably, the step of synchronously comparing the real-time collected changes in reflected light intensity based on the established dust dynamic sensing benchmark includes: The reflected light intensity is continuously and frequently acquired in real time. The incident light power is kept constant and the reflection angle and optical path length are fixed. The reflected light intensity value and time index at each sampling moment are recorded to form a continuous time series dataset of reflected light intensity. Based on the stable fluctuation characteristics in the dust dynamic sensing benchmark, the real-time reflected light intensity is synchronously compared. When the change amplitude of the reflected light intensity is consistent with the incident light intensity, the environment is determined to be stable. When the change rhythm deviates from the stable fluctuation characteristics, it is determined that there is dust obstruction and the deviation time range is recorded. For reflected light intensity data that deviates from the time range, occlusion factors are determined and spectral data are labeled. The data corresponding to the time segment is identified as environmental occlusion status, and a time index, light intensity value range, and determination status are added to the data record structure. The labeled spectral data is continuously maintained, and the dust impact status data is continuously connected with the unaffected data in the time dimension. In subsequent acquisitions, the data is continuously compared with the stable fluctuation characteristics of the dust dynamic sensing benchmark to achieve status updates.

[0010] Preferably, in the process of determining the occlusion factors and annotating the spectral data of reflected light intensity data that deviates from the time range, the spectral data annotation includes associating the time index of the dust-affected state with the light intensity value range, and synchronously generating an environmental occlusion status identifier in the data recording structure, so that the dust-affected state data maintains a continuous connection with the unaffected data in the time dimension, ensuring the integrity of the spectral data structure and continuous updating in subsequent acquisition processes.

[0011] Preferably, the step of performing trend constraint processing on the spectral data already labeled as dust-affected states includes: Using the time index recorded in the dust dynamic sensing benchmark as a reference, the spectral data group under dust influence is extracted and rearranged in chronological order. By comparing the unaffected spectral data before and after the dust-affected section, the average intensity level of elemental characteristic peaks is determined, forming the affected spectral sequence. Using the starting time of the dust impact state as the base point, the characteristic peak intensity values ​​of several sampling points before and after are selected to form a time-continuous change curve. This curve is then compared with the light intensity change range of stable fluctuation characteristics to determine the characteristic peak of abnormal sudden drop. Using stable fluctuation characteristics as a constraint benchmark, the amplitude of abnormal intensity drop caused by dust blockage is limited, and the intensity value is adjusted according to the characteristic peak intensity before and after the time, so that the change of characteristic peak intensity is included in the continuous change range. The characteristic peak intensity sequence processed by trend constraint is continuously maintained. The intensity corresponding to the start and end times of the dust-affected section is compared with the intensity change trend of the adjacent unaffected section. The intensity difference at the boundary is adjusted to maintain the smooth transition of the spectral curve in the time dimension.

[0012] Preferably, when maintaining the continuity of the characteristic peak intensity sequence after trend constraint processing, the overall intensity change range is limited based on stable fluctuation characteristics. The characteristic peak intensity corresponding to the start and end times of the dust-affected section is compared with the intensity change trend of the adjacent unaffected section, and the intensity difference is adjusted at the boundary so that the spectral curve remains smooth and continuous in the time dimension and consistent with the actual reflection change trend of the mineral sample.

[0013] Preferably, the step of continuously connecting the component change rhythms of adjacent time points based on the characteristic peak intensity sequence after trend constraint processing includes: By selecting the boundary positions between the dust-affected section and the unaffected sections before and after it, a continuous time reference is constructed. The start time, end time, and spectral data points of the dust-affected section are determined by the time index as the main line, forming a continuous index relationship in the time dimension. Trend matching of characteristic peak intensities between dust-affected sections and adjacent unaffected sections is performed. The direction of change of characteristic peak intensity at the end of the dust-affected section is compared with the initial direction of change of subsequent sections. The rate of change at the connection is adjusted according to the stable fluctuation characteristics to form a smooth transition zone. Based on the characteristic peak intensity sequence analysis after trend matching, the direction and rate of change of intensity ratio between adjacent time points are analyzed, and the data inside the dust-affected section are connected with the data of the unaffected sections on both sides in terms of compositional change dimension to limit the reasonable range of change. After completing the continuous connection of the composition change rhythm, the data of the dust-affected section and the unaffected section are integrated on the time axis, rearranged and continuously merged according to the time index to form a complete and smooth composition quantitative result sequence.

[0014] Preferably, the step of back-inputting the obtained continuous component quantification results into the dust dynamic sensing benchmark update process includes: The continuous component quantitative results are correlated with the time index of the dust dynamic sensing benchmark. By comparing the continuous component quantitative results with the time series of incident light intensity and reflected light intensity, the correspondence between the component quantitative values ​​and the light intensity change state at each moment is determined, forming a reverse reference basis for the environmental state. The reverse correlation results are input into the dust dynamic sensing benchmark update process. Using stable fluctuation characteristics as a reference, the relationship between the change range of the quantitative results of continuous components and the original light intensity fluctuation range is compared. The judgment range of the dust dynamic sensing benchmark is dynamically adjusted based on the comparison results. After the dust dynamic sensing benchmark judgment range is adjusted, the updated sensing benchmark data is connected with the benchmark data of the previous detection cycle in time. Through the transition between the old and new benchmark fluctuation ranges, a gradually evolving benchmark curve is formed to maintain the smoothness and continuity of benchmark changes. The updated sensing benchmark is combined with stable fluctuation characteristics to form a new light intensity fluctuation reference range. In subsequent detection processes, the optical path status is identified in real time based on the new judgment range, maintaining the continuity of detection data and the stability of the detection system.

[0015] The bauxite composition quantitative detection system includes a dust sensing benchmark establishment module, a reflected light intensity comparison module, a characteristic peak trend constraint module, a composition change fusion module, and a benchmark dynamic update module. Dust sensing benchmark establishment module: Establishes a dynamic dust sensing benchmark around the optical path area of ​​the intelligent sensing system. By continuously and frequently collecting incident light intensity over time, it extracts the stable fluctuation characteristics of the ambient background light intensity. Reflected light intensity comparison module: Based on the established dust dynamic sensing benchmark, the real-time collected reflected light intensity changes are synchronously compared. When the rhythm of reflected light intensity changes deviates from the stable fluctuation characteristics, it is determined that the current collected data contains environmental occlusion factors, and the corresponding spectral data is marked as the dust influence state. Characteristic Peak Trend Constraint Module: For spectral data that has been labeled as being affected by dust, the module performs trend constraint processing on the intensity changes of elemental characteristic peaks within the time period, limits the amplitude of abnormal changes based on stable fluctuation characteristics, and includes sudden attenuation within the range of continuous changes. Composition variation fusion module: Based on the characteristic peak intensity sequence after trend constraint processing, the composition variation rhythm of adjacent time moments is continuously connected, and the data corresponding to the dust-affected section is continuously fused with the data of the unaffected section to obtain the quantitative results of composition. The benchmark dynamic update module: continuously obtains quantitative results of components and inputs them back into the dust dynamic sensing benchmark update process, and dynamically adjusts the judgment range of the dust dynamic sensing benchmark in combination with stable fluctuation characteristics.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention establishes a dynamic dust sensing benchmark in the detection optical path area, creating a continuous temporal correspondence between incident light intensity and reflected light signal. This allows for real-time identification of abnormal light intensity changes caused by dust obstruction during the detection process. Through this dynamic sensing and synchronous comparison method, the detection process can accurately distinguish between environmental disturbances and changes in mineral sample reflection, effectively avoiding misjudgments caused by sudden drops in spectral signal due to dust obstruction. This ensures the stability and continuity of the detection signal, providing a reliable data foundation for subsequent bauxite composition analysis.

[0017] This invention employs trend-constrained processing and continuous fusion of dust-affected spectral data to ensure smooth and continuous changes in elemental characteristic peak intensity over time, eliminating instantaneous intensity fluctuations caused by dust obstruction. The processed quantitative results achieve dynamic feedback linkage with environmental benchmarks, enabling the dust sensing benchmark to automatically adjust its judgment range according to changes in the detection environment. This maintains detection accuracy and data consistency even under continuous dust disturbance conditions, significantly improving the stability and reliability of bauxite component detection. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of the method for quantitative detection of bauxite components according to the present invention.

[0020] Figure 2 This is a schematic diagram of the modules of the bauxite composition quantitative detection system of the present invention. Detailed Implementation

[0021] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0022] This invention provides, for example Figure 1 The method for quantitative detection of bauxite components shown includes the following steps: Step 1: Establish a dynamic dust sensing benchmark around the optical path area of ​​the intelligent sensing system. By continuously and frequently collecting incident light intensity over time, extract the stable fluctuation characteristics of the ambient background light intensity and use these stable fluctuation characteristics as a reference benchmark for identifying the dust occlusion state. The specific implementation method for this step is as follows: Before the detection device is put into operation, the geometric layout of the optical path region is accurately confirmed, including the relative positional relationship between the incident light source emitting end and the reflected light receiving end. The light source emitting end is fixed at a predetermined angle to the surface of the bauxite sample, allowing the incident light beam to propagate along a specific path in the air before reaching the sample surface. The reflected light receiving end is positioned symmetrically along the incident light path to receive the light signal reflected from the sample surface. To ensure the stability of the optical path region, the entire optical path region is kept in a closed channel state to minimize the impact of external airflow disturbances on beam propagation. Subsequently, a light intensity acquisition channel is established along the beam propagation path, and the change of incident light intensity over time is recorded using continuous high-frequency sampling. The sampling process is carried out under stable environmental conditions and low dust concentration, with the sampling frequency set in the millisecond range to fully capture the natural fluctuation characteristics of the incident light intensity. When the incident light intensity changes over time in a stable environment, it exhibits regular, subtle fluctuations. These fluctuations reflect the micro-energy changes caused by air molecular perturbations, temperature gradients, and background light interference. By continuously recording the light intensity variation curve over a relatively long period of time, the temporal distribution pattern of incident light under natural conditions is obtained, thereby forming the initial light intensity dataset of the optical path region.

[0023] After obtaining the time-series data of incident light intensity, long-term high-frequency sampling is continued on the same optical path region to cover different time segments of the detection environment. During sampling, the light source power and optical path position are kept constant to ensure that the changes reflected in the data originate solely from environmental conditions. Multiple sampling results are time-aligned to obtain a group of fluctuation curves of incident light intensity in different time intervals. From these curves, stable segments without significant interference are selected, and the maximum, minimum, and average values ​​of light intensity are statistically analyzed to determine the fluctuation range of light intensity under natural conditions. By analyzing the amplitude and period of light intensity changes at continuous time points, the fluctuation rhythm and amplitude range of incident light under stable conditions are extracted, forming a stable fluctuation characteristic of the environmental background light intensity. This characteristic reflects the normal fluctuation range of light intensity over time under dust-free conditions, providing a clear reference limit for subsequent judgment of optical path environmental anomalies. At this stage, all light intensity fluctuation data and time information are stored point-by-point, forming a continuous time-intensity dataset containing the light intensity value and corresponding time index for each sampling point.

[0024] After extracting the stable fluctuation characteristics, a dynamic dust sensing benchmark is established based on these characteristics. This benchmark is based on the time series of incident light intensity, and the stability of the optical path environment is determined by real-time monitoring of the changing trend of the incident light intensity. During implementation, incident light in the optical path region is continuously collected at the same sampling frequency as described above to ensure consistent temporal resolution. Each new light intensity sample value is compared point-by-point with the corresponding time segment in the aforementioned stable fluctuation characteristics. When the amplitude, rate, and direction of change of the sample value are all within the normal range of the stable fluctuation characteristics, the optical path environment is determined to be unobstructed. When the changing trend of the sample value exceeds the defined range of the stable fluctuation characteristics, it indicates that unnatural fluctuations have occurred in the optical path region, and dust particles may have entered the propagation path. At this time, the incident light intensity at that time point and the adjacent time segments are marked as being affected by dust for subsequent reflection signal processing. The dust dynamic sensing benchmark is formed in this process. It includes the light intensity change state of the optical path area at different times and its corresponding stable fluctuation range. Through continuous comparison, environmental disturbances in the optical path can be identified in real time, and abnormal states of light intensity change can be accurately located in time.

[0025] After establishing a dynamic dust sensing benchmark, it is continuously updated and expanded to maintain its effectiveness throughout the quantitative detection of bauxite components. During detection operations, incident light intensity is affected by a combination of environmental dust concentration, air humidity, and the surface reflectivity of the mineral sample. Therefore, continuously collected light intensity data needs to be synchronously correlated with the dynamic dust sensing benchmark. Real-time incident light intensity data is continuously incorporated into the established sensing benchmark, and combined with newly acquired light intensity fluctuation ranges, the original stable fluctuation characteristics are dynamically expanded so that the benchmark can reflect the natural changing trend of the detection environment over time. When the detection environment gradually experiences an increase in dust concentration or air disturbance, the judgment range of the dynamic dust sensing benchmark is adjusted accordingly, thereby ensuring that the sensitivity of the optical path environmental state identification remains within a reasonable range. In the continuous operation of quantitative bauxite component detection, the dynamic dust sensing benchmark always serves as a unified reference basis for identifying dust obstruction states. By continuously recording the correspondence between incident light intensity and time, the stability of the detection optical path is maintained, ensuring that the reflected light signal remains traceable and continuous even when affected by environmental fluctuations. Through this continuous time update mechanism, the dust dynamic sensing benchmark realizes the dynamic reflection of the optical path environment, ensuring that the change of incident light intensity is consistent with environmental fluctuations throughout the detection process, and providing a stable and reliable reference basis for subsequent synchronous comparison of reflected light signals and spectral data analysis.

[0026] Through the above steps, a dynamic dust sensing benchmark for the optical path region is constructed and maintained in the specific detection environment. Its establishment process is entirely based on continuous high-frequency acquisition of incident light intensity over time and extraction of stable fluctuations in ambient background light intensity, forming a light intensity reference system with temporal continuity and environmental adaptability. In the quantitative detection of bauxite components, by establishing this benchmark, the detection process can promptly identify obstruction states when dust concentration suddenly increases or fluctuates, maintaining stable transmission of the detection optical path and ensuring the consistency and continuity of the acquired reflected light data in the time dimension, providing a reliable foundation for subsequent quantitative analysis.

[0027] Step 2: Based on the established dust dynamic sensing benchmark, the real-time collected changes in reflected light intensity are synchronously compared. When the rhythm of reflected light intensity changes deviates from the stable fluctuation characteristics, it is determined that the current collected data contains environmental shading factors, and the corresponding spectral data is marked as the dust influence state. The specific implementation method for this step is as follows: After establishing the dust dynamic sensing benchmark, continuous high-frequency acquisition of reflected light intensity was performed. The reflected light signal originates from the beam of light reflected after incident light strikes the surface of the bauxite sample. This beam propagates through the air on its return path and is then received by the receiver. During the acquisition process, the incident light power, reflection angle, and optical path length were kept constant to ensure that changes in reflected light intensity only stem from the surface characteristics of the mineral sample and environmental factors. The sampling frequency of reflected light intensity was consistent with the incident light intensity sampling frequency in the dust dynamic sensing benchmark, ensuring a one-to-one correspondence between the two sets of data over time. The reflected light intensity value at each sampling moment was recorded, along with a time index, forming a continuous time-series dataset of reflected light intensity. In this way, changes in reflected light intensity can be synchronized with changes in incident light intensity over time, laying the foundation for subsequent synchronous comparison.

[0028] After obtaining the time series of real-time reflected light intensity, the real-time reflected light intensity is synchronously compared based on the stable fluctuation characteristics formed in the dust dynamic sensing benchmark. During the comparison, the incident light intensity fluctuation characteristics under the same time index are used as a reference, and the trend, amplitude, and direction of change of the reflected light intensity at that moment are compared one by one with the corresponding interval of the stable fluctuation characteristics. If the amplitude of the reflected light intensity change is relatively consistent with the amplitude of the incident light intensity change, and the rhythm of the reflected light intensity change is synchronized with the stable fluctuation characteristics of the incident light intensity, it is determined that the current optical path environment is not affected by dust, and the change of the light signal is in a normal fluctuation state. If the rhythm of the reflected light intensity change deviates from the stable fluctuation characteristics of the incident light intensity, or if there are nonlinear changes such as sudden drops or abnormal oscillations, it indicates that there may be dust particles blocking the optical path environment or non-uniform scattering caused by changes in the particle density in the air. At this time, the collected reflected light signal contains interference components caused by dust, and the trend of the reflected light intensity change is no longer consistent with the stable fluctuation characteristics of the incident light intensity.

[0029] When the rhythm of real-time reflected light intensity changes deviates from stable fluctuation characteristics, the collected data for that time period is subjected to occlusion factor determination and spectral data annotation. In this process, the starting point and duration of the deviation are first determined, and all reflected light intensity data corresponding to that time range are extracted separately. Based on the time range of the dust dynamic sensing benchmark, this data segment is marked as an environmental occlusion state. For spectral data within the deviation range, status identification information is added to the data record structure to distinguish spectral data affected by dust from normal sampling data. This identification includes three parts: time index, light intensity value range, and determination status, used for special processing of these data in subsequent processing stages. Through this annotation method, dust occlusion events can be accurately located temporally without affecting the overall data continuity, enabling the detection data to reflect the true sampling environment state.

[0030] After completing the environmental occlusion labeling of the spectral data, the continuity of the entire time series data structure is maintained. To ensure that subsequent data processing can correctly identify and utilize this labeling information, the spectral data of the dust-affected state are continuously connected with the unaffected data in the time dimension, maintaining the integrity of the data structure. In the specific implementation process, the time index is kept continuously numbered, and the dust-affected state identifier is inserted in the corresponding time segment, enabling automatic identification of the state during data reading. The dust dynamic sensing benchmark continues to serve as a reference basis at this stage, maintaining continuous perception of the optical path environment state. After new reflected light intensity data is acquired, it is again synchronously compared with the stable fluctuation characteristics of the dust dynamic sensing benchmark, thereby achieving continuous state updates. Through this continuous comparison and labeling method, a dynamic closed loop can be formed in the detection process, ensuring that the occurrence, duration, and end of dust occlusion are completely recorded at the data level. The entire process enables the quantitative detection of bauxite components to maintain data traceability and the reliability of detection results in environments with drastic changes in dust concentration.

[0031] Through the continuous implementation of the above steps, the quantitative detection of bauxite components achieves synchronous comparison of real-time reflected light intensity changes, immediate determination of environmental occlusion status, and accurate labeling of spectral data. This process fully utilizes the stable fluctuation characteristics established in the dust dynamic sensing benchmark, synchronously correlating the time-series data of incident and reflected light. This ensures that interference from environmental factors on the light signal can be identified in real time and labeled in the data, enabling subsequent component inversion and ratio calculations to be based on reliable data, thereby guaranteeing the continuity and consistency of the entire detection process in complex mining environments.

[0032] Step 3: For the spectral data that has been marked as being affected by dust, trend constraint processing is applied to the changes in the intensity of the elemental characteristic peaks within this time period. Based on the stable fluctuation characteristics, the amplitude of abnormal changes is limited, and sudden attenuation is included in the range of continuous changes, thereby suppressing the instantaneous instability of the characteristic peak intensity ratio. The specific implementation method for this step is as follows: After obtaining the spectral data marked as being under dust influence, all spectra within that time period were refined. Using the time index recorded in the dust dynamic sensing benchmark as a reference, the spectral data sets under dust influence were extracted and rearranged chronologically to form a complete affected spectral sequence. In this sequence, each spectrum corresponds to the reflected light signal and its spectral distribution characteristics at a specific sampling time. By comparing the unaffected spectral data before and after the dust-affected zone, the average intensity levels of elemental characteristic peaks before and after the dust obstruction event were determined. Since dust obstruction causes partial scattering or absorption of incident light, leading to a decrease in the overall intensity of the reflected spectrum, the sudden drop in characteristic peak intensity within this time period is often not caused by changes in mineral composition, but rather by instantaneous disturbances caused by environmental factors. By clearly defining the time range and spectral characteristic distribution of the dust-affected zone, boundary conditions can be provided for subsequent trend constraints, allowing changes in characteristic peak intensity to be adjusted within a reasonable range.

[0033] After identifying the spectral sequence of the dust-affected area, the intensity variation trend of the elemental characteristic peaks within this time period is extracted. Using the starting time marked by the dust-affected state as the base point, characteristic peak intensity values ​​from several sampling points before and after this time are selected to form a time-continuous intensity variation curve. By observing the direction, amplitude, and rate of change of characteristic peak intensity, the attenuation characteristics of the spectral signal during dust obstruction are determined. Dust obstruction-induced intensity drops typically manifest as a sharp decrease over a short period, while normal mineral sample reflectance changes exhibit a slow, continuous trend. By comparing the intensity variation trend of the dust-affected area with the light intensity variation range recorded in the stable fluctuation characteristics, it is possible to identify which characteristic peak changes constitute abnormal drops. The core of this process is to re-establish the correspondence between the spectral changes under dust influence and the normal fluctuation range, ensuring that abnormal changes are confined within the upper and lower limits defined by the stable fluctuation characteristics, thus providing a basis for trend constraint.

[0034] After extracting the characteristic peak intensity variation trend, trend constraint processing was applied to the spectral data within the dust-affected area. Using stable fluctuation characteristics as the constraint benchmark, the amplitude of abrupt intensity drops caused by dust obstruction was limited according to the fluctuation range. Specifically, when the characteristic peak intensity at a certain time point drops beyond the normal lower limit of stable fluctuation characteristics, this drop is no longer considered a change in the mineral sample's reflectance performance. Instead, the amplitude is adjusted based on the characteristic peak intensities at previous and subsequent times, ensuring that the intensity value at that moment falls within a continuous variation range. In this way, sudden attenuation no longer manifests as a single sharp fluctuation but is smoothly integrated into the overall trend of spectral intensity variation. During the processing, the consistency between the direction of characteristic peak intensity variation and the time evolution trend is maintained to ensure the continuity of the spectrum over time. Through trend constraint, the instantaneous abnormal fluctuations caused by dust obstruction are effectively suppressed, and the spectral intensity curve is restored to a continuous variation state consistent with the actual surface reflectance characteristics of the mineral sample.

[0035] After completing the trend constraint processing, the continuity of the constrained characteristic peak intensity sequence is maintained, allowing the adjusted spectral data to seamlessly connect to the spectral segments unaffected by dust. By comparing the characteristic peak intensities corresponding to the start and end times of the dust-affected segment with the intensity change trends of the adjacent unaffected segments, the intensity difference at the boundary is adjusted to ensure a smooth transition of the spectral curves at the connection points. During this process, the overall intensity change range is further limited based on stable fluctuation characteristics, ensuring that the spectral intensity of the entire time series remains within a reasonable fluctuation range. After continuity maintenance, the spectral data within the dust-affected segment no longer exhibits abrupt attenuation characteristics but naturally transitions to the trend of true reflectance changes in the mineral sample. The final characteristic peak intensity sequence retains the true reflectance differences caused by changes in mineral composition while eliminating instantaneous anomalies caused by dust interference, maintaining the consistency and stability of the spectral data over time and providing a reliable foundation for subsequent quantitative component analysis.

[0036] Through the above implementation steps, trend constraint processing was completed on the spectral data already marked as being affected by dust. The entire process uses stable fluctuation characteristics as a reference, limiting the amplitude of abnormal changes to incorporate sudden attenuation caused by dust obstruction into the continuous spectral variation range. This effectively suppresses the instantaneous instability of the intensity ratio of elemental characteristic peaks, ensuring that the spectral data in the quantitative detection of bauxite components remains smooth and continuous, guaranteeing the stable reliability and comparability of the detection results even under dust disturbance environments.

[0037] Step 4: Based on the characteristic peak intensity sequence after trend constraint processing, the component change rhythm of adjacent time points is continuously connected, and the data corresponding to the dust-affected section is continuously fused with the data of the unaffected section to obtain consistent component quantification results in the time dimension. The specific implementation method for this step is as follows: After obtaining the characteristic peak intensity sequence after trend constraint processing, the boundary positions of the dust-affected section and the unaffected sections before and after it are selected to construct a continuous time reference. Using the time index as the main line, the start and end times of the dust-affected section, as well as the spectral data points of adjacent preceding and subsequent times, are determined. For the spectral sequence within the dust-affected section, the abnormal attenuation has been smoothed through trend constraint processing; therefore, the characteristic peak intensity changes within this section can be considered as fluctuation curves closely resembling the actual reflectance characteristics of the mineral sample. By comparing the characteristic peak intensity at the end of the dust-affected section with the characteristic peak intensity at the start of the subsequent unaffected section, the intensity difference and trend between the two are observed to determine the smooth transition direction when connecting the two data segments. The main task of this stage is to form a continuous index relationship in the time dimension, providing a time reference framework for the subsequent connection of component change rhythms, so that the characteristic peak intensity at each time moment can find corresponding upper and lower correlation points on the time axis.

[0038] After establishing a clear time baseline, trend matching is performed on the characteristic peak intensities between the dust-affected section and the adjacent unaffected section. Using the characteristic peak intensity at the end of the dust-affected section as a baseline, its direction of change is compared with the initial direction of change in the subsequent section. When the two data segments have the same direction of change and similar rates of intensity change, they can be directly connected sequentially in time, allowing the characteristic peak intensity curve to extend naturally. When there is a deviation in the direction of change or a significant difference in the rate of change between the two data segments, the rate of change at the connection point is adjusted based on the stable fluctuation characteristics formed in the previous stage, ensuring it conforms to the continuous law of overall spectral fluctuation. Through this trend matching process, a smooth transition zone is formed at the connection point between the dust-affected section and the unaffected section, thus eliminating the intensity jump phenomenon caused by sudden environmental changes. Simultaneously, the same trend matching is performed on the characteristic peak intensities between the beginning of the dust-affected section and the preceding unaffected section, ensuring that the entire dust-affected section maintains a smooth connection with normal detection data at both ends of time, guaranteeing the temporal continuity and physical consistency of the data.

[0039] After matching the characteristic peak intensity trends, a continuous connection of the component variation rhythm is performed. The component variation rhythm reflects the dynamic proportional relationship of the major components in the mineral sample over time. Based on the characteristic peak intensity sequence after trend matching, the direction and rate of change of the intensity ratio between adjacent time points are analyzed, and the data within the dust-affected section are connected with the data of the unaffected sections on both sides in the dimension of component variation. Specifically, the characteristic peak intensity ratio at the end of the dust-affected section is smoothly connected with the ratio at subsequent time points, so that the component variation curve maintains a continuous variation trend before and after the dust-covered section. The characteristic peak intensity ratio at the beginning of the dust-affected section is also processed to ensure that it maintains a consistent variation direction with the data of the preceding unaffected section. Throughout the connection process, a reasonable range of component variation amplitude is limited by stable fluctuation characteristics to ensure that the intensity adjustment formed in the dust-affected section does not cause abnormal jumps in component proportions. Through this continuous connection, the component variation curves in the dust-affected section and the variation curves in the unaffected section are seamlessly integrated in the time dimension, so that the mineral sample components show a continuous and smooth variation trend in the detection time series.

[0040] After completing the continuous connection of the compositional change rhythm, the data from the entire detection time segment are fused and output uniformly. Data from dust-affected areas are integrated with data from unaffected areas on the time axis to form a complete sequence of quantitative composition results. By re-analyzing the intensity change trend of the entire time series, the characteristic peak intensities after trend constraint processing are kept consistent across the time dimension. The characteristic peak intensity ratios of each element are rearranged and continuously fused according to the time index, ensuring no abrupt changes or discontinuities between dust-affected and normal areas. The fused quantitative composition results reflect the true trend of mineral sample composition changes under dust disturbance conditions, maintaining the stability of the detection results under long-term continuous operation. Through this continuous fusion, data from dust-affected areas are naturally integrated into the overall detection sequence, preserving the dynamic characteristics of mineral sample composition changes over time while eliminating instantaneous abnormal fluctuations caused by dust obstruction, enabling a smooth transition of the entire detection process across the time dimension.

[0041] Through the above steps, based on the characteristic peak intensity sequence after trend constraint processing, a continuous connection of the composition change rhythm between adjacent time points was achieved. Data from the dust-affected section and the unaffected section were fused temporally, resulting in consistent quantitative composition results over time. This process not only restored the spectral continuity disrupted by dust interference but also ensured the stability and continuity of the composition change curve throughout the entire detection process, providing complete and reliable basic data support for subsequent quantitative analysis, result presentation, and smelting batch optimization.

[0042] Step 5: Input the obtained continuous component quantitative results back into the dust dynamic sensing benchmark update process, and dynamically adjust the judgment range of the dust dynamic sensing benchmark in combination with the stable fluctuation characteristics, so as to maintain the quantitative detection accuracy and stability of the detection system under continuous dust disturbance conditions. The specific implementation method for this step is as follows: After obtaining the continuous component quantitative results, these results are correlated with the time index of the dust dynamic sensing benchmark. By comparing the continuous component quantitative results with the time series of incident and reflected light intensities, the correspondence between the component quantitative value and the corresponding light intensity change state at each moment within the detection period is determined. Since the continuous component quantitative results are formed after dust influence identification, trend constraint, and continuous connection of component rhythm, their change curves reflect the true component change trend of the mineral sample during actual detection, while the small fluctuations may reflect the residual influence of dust disturbance on the optical path signal. Through this time correspondence, the component quantitative results can be used as a reverse reference basis for the environmental state, identifying moments in a specific time period where the light intensity change and component change are inconsistent. These moments often correspond to the range of weak influence of dust disturbance on the optical path. This step provides direct feedback for the dynamic updating of the dust dynamic sensing benchmark, enabling the component quantitative results to participate in the self-correction of the sensing benchmark.

[0043] After establishing the correlation between the continuous component quantitative results and the time index, the reverse correlation results are input into the dust dynamic sensing benchmark update process. Using stable fluctuation characteristics as a reference, the relationship between the variation amplitude of the continuous component quantitative results at each time point and the light intensity fluctuation range in the original dust dynamic sensing benchmark is compared. When the continuous component quantitative results remain stable within a certain period, but the light intensity fluctuation exceeds the normal range defined by the stable fluctuation characteristics, it indicates that the judgment interval of the dust dynamic sensing benchmark is too tight, and the upper limit of fluctuation needs to be expanded. Conversely, when the continuous component quantitative results exhibit unstable fluctuations within a certain period, but the light intensity fluctuation remains within the original stable range, it indicates that the current benchmark judgment interval is too wide, and the lower limit of fluctuation needs to be narrowed. In this way, the judgment range of the dust dynamic sensing benchmark is no longer a fixed interval, but is dynamically adjusted according to the stability of the mineral sample composition change trend during the detection process. This step realizes the reverse information flow from the detection results to the environmental benchmark, enabling the dust dynamic sensing benchmark to maintain a sensitive and robust balance in a continuously changing detection environment.

[0044] After initial dynamic adjustment of the judgment range of the dust dynamic sensing benchmark, the updated benchmark is continuously extended over time to maintain its long-term stable reference function throughout the detection process. Specifically, the adjusted benchmark data is time-linked with the benchmark data from the previous detection cycle to maintain the smoothness and continuity of benchmark changes. By transitioning the fluctuation ranges of the old and new benchmarks, a benchmark curve that evolves gradually over time is formed, enabling the dust dynamic sensing benchmark to reflect the long-term trend and short-term fluctuation characteristics of dust concentration. Simultaneously, subtle changes in continuous component quantification results are continuously fed back to the sensing benchmark, forming a continuous dynamic calibration mechanism. When the dust concentration in the detection environment periodically increases or decreases, the upper and lower limits of the judgment range of the dust dynamic sensing benchmark are adjusted accordingly, ensuring that the light intensity fluctuation characteristics and component change curves remain consistent over time. This process ensures that the dust dynamic sensing benchmark will not deviate excessively due to a single disturbance during continuous operation, thus maintaining the long-term stability of the sensing results.

[0045] After the dust dynamic sensing benchmark undergoes reverse input and dynamic adjustment, an environmentally adaptive fusion is performed on the updated benchmark to ensure that it can maintain the accuracy and stability of quantitative detection under continuous dust disturbance conditions. By combining the updated sensing benchmark with stable fluctuation characteristics, a new light intensity fluctuation reference range is formed, so that the judgment criteria of the sensing benchmark can reflect changes in the real-time detection environment while retaining the natural fluctuation law of the optical path represented by the stable fluctuation characteristics. In subsequent detection processes, the dust dynamic sensing benchmark uses the new judgment range as a reference basis to instantly identify the optical path state during each incident and reflected light acquisition. When the detection environment experiences continuous dust disturbance, the updated sensing benchmark can adjust the judgment threshold for light intensity fluctuations in a timely manner, avoiding misjudging normal fluctuations as abnormal occlusion, thereby maintaining the continuity and accuracy of detection data. At the same time, when the environment returns to stability, the sensing benchmark can automatically return to the standard range defined by the stable fluctuation characteristics, enabling the entire detection system to achieve a balance between dynamic and stable states. In this process, the quantitative results of continuous components not only exist as detection output data but also serve as a feedback source for updating the sensing benchmark, giving the detection system the ability to self-adapt and continuously correct itself.

[0046] Through the above steps, the obtained continuous quantitative results of components are fed back into the updating process of the dust dynamic sensing benchmark. Combined with stable fluctuation characteristics, the judgment range of the dust dynamic sensing benchmark is dynamically adjusted, achieving a dual maintenance of detection accuracy and system stability under continuous dust disturbance conditions. The entire process constitutes a closed-loop detection and feedback system, enabling the dust dynamic sensing benchmark to be continuously optimized based on actual detection results. This achieves high-precision quantitative component detection and long-term operational stability in complex mining environments, providing reliable technical support for the intelligent and automated detection of bauxite components.

[0047] This invention establishes a dynamic dust sensing benchmark in the detection optical path area, creating a continuous temporal correspondence between incident light intensity and reflected light signal. This allows for real-time identification of abnormal light intensity changes caused by dust obstruction during the detection process. Through this dynamic sensing and synchronous comparison method, the detection process can accurately distinguish between environmental disturbances and changes in mineral sample reflection, effectively avoiding misjudgments caused by sudden drops in spectral signal due to dust obstruction. This ensures the stability and continuity of the detection signal, providing a reliable data foundation for subsequent bauxite composition analysis.

[0048] This invention employs trend-constrained processing and continuous fusion of dust-affected spectral data to ensure smooth and continuous changes in elemental characteristic peak intensity over time, eliminating instantaneous intensity fluctuations caused by dust obstruction. The processed quantitative results achieve dynamic feedback linkage with environmental benchmarks, enabling the dust sensing benchmark to automatically adjust its judgment range according to changes in the detection environment. This maintains detection accuracy and data consistency even under continuous dust disturbance conditions, significantly improving the stability and reliability of bauxite component detection.

[0049] This invention provides, for example Figure 2 The bauxite composition quantitative detection system shown includes a dust sensing benchmark establishment module, a reflected light intensity comparison module, a characteristic peak trend constraint module, a composition change fusion module, and a benchmark dynamic update module. Dust sensing benchmark establishment module: Establishes a dynamic dust sensing benchmark around the optical path area of ​​the intelligent sensing system. By continuously and frequently collecting incident light intensity over time, it extracts the stable fluctuation characteristics of the ambient background light intensity. Reflected light intensity comparison module: Based on the established dust dynamic sensing benchmark, the real-time collected reflected light intensity changes are synchronously compared. When the rhythm of reflected light intensity changes deviates from the stable fluctuation characteristics, it is determined that the current collected data contains environmental occlusion factors, and the corresponding spectral data is marked as the dust influence state. Characteristic Peak Trend Constraint Module: For spectral data that has been labeled as being affected by dust, the module performs trend constraint processing on the intensity changes of elemental characteristic peaks within the time period, limits the amplitude of abnormal changes based on stable fluctuation characteristics, and includes sudden attenuation within the range of continuous changes. Composition variation fusion module: Based on the characteristic peak intensity sequence after trend constraint processing, the composition variation rhythm of adjacent time moments is continuously connected, and the data corresponding to the dust-affected section is continuously fused with the data of the unaffected section to obtain the quantitative results of composition. The benchmark dynamic update module: continuously obtains quantitative results of components and inputs them back into the dust dynamic sensing benchmark update process, and dynamically adjusts the judgment range of the dust dynamic sensing benchmark in combination with stable fluctuation characteristics.

[0050] The method for quantitative detection of bauxite components provided in this embodiment of the invention is implemented by the above-mentioned bauxite component quantitative detection system. The specific methods and procedures of the bauxite component quantitative detection system are detailed in the embodiments of the above-mentioned method for quantitative detection of bauxite components, and will not be repeated here.

[0051] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for quantitative detection of bauxite components, characterized in that, Includes the following steps: Step 1: Establish a dynamic dust sensing benchmark around the optical path area of ​​the intelligent sensing system. Extract the stable fluctuation characteristics of the ambient background light intensity by continuously and frequently collecting the incident light intensity over time. Step 2: Based on the established dust dynamic sensing benchmark, the real-time collected changes in reflected light intensity are synchronously compared. When the rhythm of reflected light intensity changes deviates from the stable fluctuation characteristics, it is determined that the current collected data contains environmental shading factors, and the corresponding spectral data is marked as the dust influence state. Step 3: For the spectral data that has been marked as being affected by dust, perform trend constraint processing on the changes in the intensity of the elemental characteristic peaks within this time period, limit the amplitude of abnormal changes based on the stable fluctuation characteristics, and include sudden attenuation in the range of continuous changes. Step 4: Based on the characteristic peak intensity sequence after trend constraint processing, the component change rhythm of adjacent time points is continuously connected, and the data corresponding to the dust-affected section is continuously fused with the data of the unaffected section to obtain the component quantitative results; Step 5: Input the continuously obtained quantitative results of components back into the dust dynamic sensing benchmark update process, and dynamically adjust the judgment range of the dust dynamic sensing benchmark in combination with stable fluctuation characteristics.

2. The method for quantitative detection of bauxite components according to claim 1, characterized in that, The steps for establishing a dynamic dust sensing baseline around the optical path area of ​​an intelligent sensing system include: Accurately confirm the geometric layout of the optical path region, determine the relative positional relationship between the incident light source emitting end and the reflected light receiving end, and establish a light intensity acquisition channel on the beam propagation path. Record the change of incident light intensity over time through continuous high-frequency sampling to form an initial light intensity dataset for the optical path region. Long-term high-frequency sampling of incident light intensity is performed, the sampling results are time-aligned, stable segments are selected and the maximum, minimum and average values ​​of light intensity are statistically analyzed, and the stable fluctuation characteristics of ambient background light intensity are extracted to form a continuous time light intensity dataset. A dynamic dust sensing benchmark is established based on stable fluctuation characteristics. The incident light intensity is continuously collected and the sampled values ​​are compared with the range of change of stable fluctuation characteristics point by point. When the sampled value exceeds the limit range, it is marked as a dust-affected state, forming a judgment result of the optical path environment state. Real-time incident light intensity data is incorporated into the dynamic dust sensing benchmark, and the stable fluctuation characteristics are dynamically expanded by combining the newly acquired light intensity fluctuation range.

3. The method for quantitative detection of bauxite components according to claim 2, characterized in that, During the dynamic expansion of the dust dynamic sensing benchmark, real-time incident light intensity data and stable fluctuation characteristics are correlated according to time index, and the judgment range of light intensity fluctuation range is adjusted according to the change of dust concentration in the detection environment.

4. The method for quantitative detection of bauxite components according to claim 2, characterized in that, The steps for synchronously comparing the real-time collected changes in reflected light intensity based on the established dust dynamic sensing benchmark include: The reflected light intensity is continuously and frequently acquired in real time. The incident light power is kept constant and the reflection angle and optical path length are fixed. The reflected light intensity value and time index at each sampling moment are recorded to form a continuous time series dataset of reflected light intensity. Based on the stable fluctuation characteristics in the dust dynamic sensing benchmark, the real-time reflected light intensity is synchronously compared. When the change amplitude of the reflected light intensity is consistent with the incident light intensity, the environment is determined to be stable. When the change rhythm deviates from the stable fluctuation characteristics, it is determined that there is dust obstruction and the deviation time range is recorded. For reflected light intensity data that deviates from the time range, occlusion factors are determined and spectral data are labeled. The data corresponding to the time segment is identified as environmental occlusion status, and a time index, light intensity value range, and determination status are added to the data record structure. The labeled spectral data is continuously maintained, and the dust impact status data is continuously connected with the unaffected data in the time dimension. In subsequent acquisitions, the data is continuously compared with the stable fluctuation characteristics of the dust dynamic sensing benchmark to achieve status updates.

5. The method for quantitative detection of bauxite components according to claim 4, characterized in that, In the process of determining the occlusion factors and annotating the spectral data of reflected light intensity data that deviates from the time range, the spectral data annotation includes associating the time index of the dust-affected state with the light intensity value range, and synchronously generating an environmental occlusion status identifier in the data record structure.

6. The method for quantitative detection of bauxite components according to claim 4, characterized in that, The steps for trend constraint processing of spectral data already labeled as dust-affected states include: Using the time index recorded in the dust dynamic sensing benchmark as a reference, the spectral data group under dust influence is extracted and rearranged in chronological order. By comparing the unaffected spectral data before and after the dust-affected section, the average intensity level of elemental characteristic peaks is determined, forming the affected spectral sequence. Using the starting time of the dust impact state as the base point, the characteristic peak intensity values ​​of several sampling points before and after are selected to form a time-continuous change curve. This change curve is then compared with the light intensity change range of stable fluctuation characteristics to determine the characteristic peak of the abnormal drop. Using stable fluctuation characteristics as a constraint benchmark, the amplitude of abnormal intensity drop caused by dust blockage is limited, and the intensity value is adjusted according to the characteristic peak intensity before and after the time, so that the change of characteristic peak intensity is included in the continuous change range. The characteristic peak intensity sequence processed by trend constraint is continuously maintained. The intensity corresponding to the start and end times of the dust-affected section is compared with the intensity change trend of the adjacent unaffected section. The intensity difference at the boundary is adjusted to maintain the smooth transition of the spectral curve in the time dimension.

7. The method for quantitative detection of bauxite composition according to claim 6, characterized in that, When maintaining the continuity of the characteristic peak intensity sequence after trend constraint processing, the overall intensity change range is limited based on the stable fluctuation characteristics. The characteristic peak intensity corresponding to the start and end times of the dust-affected section is compared with the intensity change trend of the adjacent unaffected section, and the intensity difference is adjusted at the boundary to keep the spectral curve smooth and continuous in the time dimension and consistent with the actual reflection change trend of the mineral sample.

8. The method for quantitative detection of bauxite components according to claim 7, characterized in that, The steps for continuously connecting the component variation rhythms of adjacent time points based on the characteristic peak intensity sequence after trend constraint processing include: By selecting the boundary positions between the dust-affected section and the unaffected sections before and after it, a continuous time reference is constructed. The start time, end time, and spectral data points of the dust-affected section are determined by the time index as the main line, forming a continuous index relationship in the time dimension. Trend matching of characteristic peak intensities between dust-affected sections and adjacent unaffected sections is performed. The direction of change of characteristic peak intensity at the end of the dust-affected section is compared with the initial direction of change of subsequent sections. The rate of change at the connection is adjusted according to the stable fluctuation characteristics to form a smooth transition zone. Based on the characteristic peak intensity sequence analysis after trend matching, the direction and rate of change of intensity ratio between adjacent time points are analyzed, and the data inside the dust-affected section are connected with the data of the unaffected sections on both sides in terms of compositional change dimension to limit the range of change. After completing the continuous connection of the composition change rhythm, the data of the dust-affected section and the data of the unaffected section are integrated on the time axis, rearranged and continuously merged according to the time index to form a sequence of composition quantitative results.

9. The method for quantitative detection of bauxite composition according to claim 8, characterized in that, The steps involved in back-inputting the obtained continuous component quantification results into the dust dynamic sensing benchmark update process include: The continuous component quantitative results are correlated with the time index of the dust dynamic sensing benchmark. By comparing the continuous component quantitative results with the time series of incident light intensity and reflected light intensity, the correspondence between the component quantitative values ​​and the light intensity change state at each moment is determined, forming a reverse reference basis for the environmental state. The reverse correlation results are input into the dust dynamic sensing benchmark update process. Using stable fluctuation characteristics as a reference, the relationship between the change range of the quantitative results of continuous components and the original light intensity fluctuation range is compared. The judgment range of the dust dynamic sensing benchmark is dynamically adjusted based on the comparison results. After the dust dynamic sensing benchmark judgment range is adjusted, the updated sensing benchmark data is connected with the benchmark data of the previous detection cycle in time, and a gradually evolving benchmark curve is formed through the transition between the old and new benchmark fluctuation ranges. The updated perception benchmark is combined with stable fluctuation characteristics to form a new light intensity fluctuation reference range. In subsequent detection processes, the optical path status is identified in real time based on the new judgment range.

10. A bauxite composition quantitative detection system, used to implement the bauxite composition quantitative detection method according to any one of claims 1-9, characterized in that, It includes a dust sensing benchmark establishment module, a reflected light intensity comparison module, a characteristic peak trend constraint module, a composition change fusion module, and a benchmark dynamic update module; Dust sensing benchmark establishment module: Establishes a dynamic dust sensing benchmark around the optical path area of ​​the intelligent sensing system. By continuously and frequently collecting incident light intensity over time, it extracts the stable fluctuation characteristics of the ambient background light intensity. Reflected light intensity comparison module: Based on the established dust dynamic sensing benchmark, the real-time collected reflected light intensity changes are synchronously compared. When the rhythm of reflected light intensity changes deviates from the stable fluctuation characteristics, it is determined that the current collected data contains environmental occlusion factors, and the corresponding spectral data is marked as the dust influence state. Characteristic Peak Trend Constraint Module: For spectral data that has been labeled as being affected by dust, the module performs trend constraint processing on the intensity changes of elemental characteristic peaks within the time period, limits the amplitude of abnormal changes based on stable fluctuation characteristics, and includes sudden attenuation within the range of continuous changes. Composition variation fusion module: Based on the characteristic peak intensity sequence after trend constraint processing, the composition variation rhythm of adjacent time moments is continuously connected, and the data corresponding to the dust-affected section is continuously fused with the data of the unaffected section to obtain the quantitative results of composition. The benchmark dynamic update module: continuously obtains quantitative results of components and inputs them back into the dust dynamic sensing benchmark update process, and dynamically adjusts the judgment range of the dust dynamic sensing benchmark in combination with stable fluctuation characteristics.

Images