Real-time detection method for mixer powder drying based on humidity sensor

Abstract

The application relates to the technical field of control science, and provides a mixing machine powder drying real-time detection method based on a humidity sensor. The method comprises the following steps: monitoring humidity data of an inner wall of a mixing machine in real time based on a humidity sensor, and constructing a data twin model; evaluating a humidity level, and judging whether the drying degree meets a threshold value; if not, compensating for efficiency by using historical data, determining parameters and credibility, simulating humidity rules, constructing an optimization space, optimizing parameters, and optimizing control to realize powder drying. The application solves the technical problem that in a powder drying scene where different material types change or process conditions change, powder drying control optimization cannot be quickly debugged and determined, leading to low powder drying quality stability, realizes real-time monitoring of humidity conditions in the mixing machine, automatically adapts to different drying requirements, does not need repeated experiments and debugging for each type of material, improves the controllability of powder drying, and guarantees the technical effect of powder drying quality.

Description

Technical Field

[0001] This application relates to the field of control science and technology, specifically to the field of intelligent control technology, and in particular to a real-time detection method for powder drying in mixers based on humidity sensors. Background Technology

[0002] With the continuous development and innovation of modern industrial technology, powder drying in mixers, as a key link in many production processes, has a crucial impact on product quality and production efficiency in terms of control accuracy and efficiency. However, existing drying detection methods optimize powder drying control by directly using empirical functions combined with the color and texture of the powder material in the mixer. While this can achieve intelligent control of powder that has already been dried in large quantities, if the material type changes or the process conditions are altered, the original empirical rules may no longer be applicable, requiring long-term drying adjustments. Summary of the Invention

[0003] This application provides a real-time detection method for powder drying in mixers based on humidity sensors. It aims to solve the technical problem of low powder drying quality stability caused by the inability to quickly debug and determine powder drying control optimization in powder drying scenarios where different material types or process conditions change.

[0004] In view of the above problems, this application provides a real-time detection method for powder drying in mixers based on a humidity sensor.

[0005] This application discloses a real-time detection method for powder drying in a mixer based on a humidity sensor. The method includes: real-time monitoring of the inner wall of the target mixer container using a humidity sensor to acquire real-time monitoring data, including humidity data collected at M key locations on the inner wall of the target mixer; acquiring the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, and constructing a data twin model of the mixer; evaluating the humidity level of the target mixer based on the changing trend of the real-time monitoring data, determining the degree of powder drying based on the humidity level, and comparing it with a preset powder drying threshold; if the degree of powder drying does not meet the preset powder drying threshold, performing efficiency compensation based on historical monitoring data of the humidity sensor, combined with the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, and determining efficiency compensation parameters and compensation reliability; simulating the humidity level patterns under different operating conditions based on the mixer data twin model, constructing a drying optimization space, and using the efficiency compensation parameters and compensation reliability as optimization auxiliary variables of the drying optimization space; performing parameter optimization in the drying optimization space to obtain the optimal combination of process parameters, and optimizing the powder drying control of the target mixer.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The aforementioned real-time detection method for powder drying in a mixer based on a humidity sensor collects humidity data from multiple key locations by monitoring the humidity of the inner wall of the target mixer container in real time. This data not only reflects the humidity status inside the mixer but also provides fundamental information for optimizing the drying process. Subsequently, by understanding the monitoring characteristics of the humidity sensor and the operating characteristics of the mixer, a data twin model of the mixer is constructed. This model can simulate the actual operating state of the mixer, helping to more accurately understand the relationship between humidity changes and the degree of powder drying. Then, using the real-time monitoring data, the humidity level inside the mixer is evaluated, and the degree of powder drying is determined accordingly. If the degree of drying does not reach a preset threshold, it indicates that the drying process needs to be adjusted. At this point, efficiency compensation is performed by combining historical data from the humidity sensor with the characteristic indicators of the mixer and the sensor, determining the compensation parameters and reliability. Then, the data twin model is used to simulate the humidity change patterns under different operating conditions, constructing a drying optimization space. In the drying optimization space, the efficiency compensation parameters and compensation reliability are used as optimization variables, and the optimal combination of process parameters is found through parameter optimization. This method achieves precise control over the powder drying process in a mixer through real-time monitoring, modeling, evaluation, and optimization, thereby improving drying efficiency and ensuring product quality.

[0007] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic flowchart of a real-time detection method for powder drying in a mixer based on a humidity sensor, as shown in one embodiment. Detailed Implementation

[0010] This application provides a real-time detection method for powder drying in mixers based on a humidity sensor. This solves the technical problem that powder drying quality stability is low because it is impossible to quickly debug and determine powder drying control optimization in powder drying scenarios where different material types or process conditions change.

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0012] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0013] Example 1 like Figure 1 As shown, this application provides a real-time detection method for powder drying in a mixer based on a humidity sensor, the method comprising: Based on a humidity sensor, the inner wall of the target mixer is monitored in real time to obtain real-time monitoring data, which includes humidity data collected at M key locations on the inner wall of the target mixer. With the continuous advancement of technology and the expansion of industrial production scale, powder drying, as a key link in many production processes, plays a crucial role in the stability of product quality and the improvement of production efficiency through its control precision and efficiency.

[0014] In this embodiment, the system terminal monitors the humidity of the inner wall of the target mixer container in real time using a humidity sensor. During the monitoring process, the humidity sensor collects humidity data at M key locations on the inner wall of the mixer, providing the system terminal with real-time monitoring data on the internal humidity of the mixer. Here, M is an integer greater than 0. The acquired real-time monitoring data not only reflects the humidity changes inside the mixer but also provides the system terminal with an important basis for precise control of the powder drying degree. Through data acquisition, the operating status of the mixer can be understood more accurately, ensuring the effectiveness and efficiency of the powder drying process.

[0015] The monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer are obtained, and a data twin model of the mixer is constructed. In one embodiment, to more accurately monitor the state of the powder drying process in the mixer, the system terminal calibrates the humidity sensor to ensure accurate measurements. Through testing, key parameters such as the sensor's measurement range, accuracy, and response time are determined. These parameters are then compiled to obtain the humidity sensor's monitoring performance indicators, reflecting its performance in real-time monitoring. Simultaneously, real-time monitoring data, such as rotational speed, power consumption, and temperature distribution, are processed and analyzed to extract information on the mixer's operating status, efficiency, and energy consumption. This extracted information is then compiled to obtain the target mixer's operational performance indicators. After obtaining the monitoring and operational performance indicators, the system terminal uses these indicators to determine the physical principles underlying humidity changes within the mixer, i.e., determining whether it's moisture diffusion or humid airflow. Subsequently, based on these physical principles, fundamental equations describing humidity changes within the mixer are established. These equations are based on partial differential equations or algebraic equations, with the choice depending on the complexity and accuracy requirements of the actual situation. Finally, the boundary conditions within the mixer are analyzed, including those related to temperature, humidity, and airflow. These boundary conditions are determined by operational performance indicators. Then, the determined basic equations and boundary conditions are integrated into a set of model equations to describe the overall process of humidity change within the mixer, thus constructing an initial mixer data twin model. Further, the system terminal uses the humidity sensor's monitoring performance indicators to compare and calibrate the actual monitoring data with the model output to ensure consistency. If the outputs are consistent, the initial mixer data twin model is directly used as the actual mixer data twin model. Otherwise, the basic equations or boundary conditions are reset until the outputs are consistent. The mixer data twin model can simulate the actual operating conditions of the mixer, providing an important basis for subsequent drying control optimization. Through this step, the actual situation of the powder drying process in the mixer can be more accurately grasped, laying a solid foundation for improving drying efficiency and quality.

[0016] Based on the changing trend of the real-time monitoring data, the humidity level corresponding to the target mixer is evaluated, the degree of powder drying is determined according to the humidity level, and compared with the preset powder drying threshold. In one embodiment, the system terminal performs time-series processing on real-time monitoring data and sets humidity evaluation features based on the sequenced data. Subsequently, it compares the set humidity evaluation features with the internal humidity levels of different humidity ranges to assess the current humidity level within the mixer. The humidity level helps the system terminal determine the degree of powder dryness, i.e., whether the moisture content in the powder meets the expected standard. Then, the humidity judgment result is compared with a preset powder drying threshold to analyze whether the current dryness meets production requirements. This step is crucial for ensuring powder drying quality; through real-time comparison and adjustment, precise control of the powder drying process can be achieved.

[0017] Furthermore, this application provides a method for assessing the humidity level corresponding to the target mixer based on the changing trend of the real-time monitoring data, and determining the degree of powder dryness based on the humidity level. The method further includes: Based on the M key locations on the inner wall, M monitoring time series data streams are determined; By analyzing the changes in the smallest humidity unit through the M monitoring time series data streams, a first humidity evaluation feature is set, which includes a drying rate feature and a humidity change sensitivity feature. Preferably, the system terminal determines M key locations on the inner wall based on the mixer's working principle, powder flow characteristics, and expected humidity distribution patterns. The selection of these locations should ensure a comprehensive reflection of the humidity conditions in different areas inside the mixer. Subsequently, real-time monitoring data from the determined M key inner wall locations is collected, and the raw humidity data is cleaned to extract outliers, improving data quality and consistency and providing a reliable data foundation for subsequent humidity analysis. After preprocessing, the humidity data from each key location is arranged according to time series, forming M monitoring time series data streams. Each data stream represents the humidity change over time at a key location, providing crucial data support for subsequent analysis and optimization. Then, a detailed analysis of the M monitoring time series data streams is performed, with particular attention paid to the relationship between the M monitoring time series data streams and the minimum humidity unit. The minimum humidity unit, determined by the accuracy of the humidity sensor, represents the smallest observable change in humidity. During the analysis, the system terminal calculates the humidity fluctuation frequency and amplitude, and sets the first humidity evaluation feature based on the calculation results. The drying rate characteristic describes how quickly humidity changes over time, helping to understand the drying efficiency of powder within the mixer. The humidity change sensitivity characteristic reflects the sensitivity to humidity changes within a small unit, which is crucial for precisely controlling the drying process and avoiding over- or under-drying. Analyzing these two characteristics provides a more comprehensive and in-depth understanding of humidity changes within the mixer's inner wall, offering strong support for optimizing the subsequent drying process.

[0018] Based on the M monitoring time series data streams, the humidity change of the container inner wall at the same moment is analyzed, and a second humidity evaluation feature is set. The second humidity evaluation feature includes relative humidity fluctuation feature and humidity distribution feature. Preferably, the system terminal performs simultaneous humidity change analysis on the container's inner wall based on monitoring time-series data streams acquired from M key locations. The core of this simultaneous humidity change analysis lies in comparing humidity data at different locations at the same time point, thereby revealing the spatial distribution and variation characteristics of humidity on the container's inner wall. In this process, the system terminal first splits the M monitoring time-series data streams to obtain humidity data sets at N time points. Subsequently, it performs data change analysis based on the humidity data sets at the N time points and sets a second humidity evaluation feature according to the analysis results. The second humidity evaluation feature includes relative humidity fluctuation characteristics and humidity distribution characteristics. Relative humidity fluctuation characteristics mainly reflect the degree and rate of change of humidity between different locations at the same time, which helps to understand the uniformity and stability of humidity inside the container. Humidity distribution characteristics describe the spatial distribution pattern of humidity on the container's inner wall, such as whether there are obvious humidity gradients or localized high-humidity areas, which is crucial for optimizing the drying process, avoiding humidity accumulation, and ensuring product quality. Through the analysis of these two features, a more comprehensive understanding of the dynamic changes in humidity on the container's inner wall can be obtained, providing strong support for the formulation of subsequent drying control strategies. For example, if it is found that the relative humidity fluctuates greatly or the humidity distribution is uneven in certain areas, the drying parameters can be adjusted or the operation mode of the mixer can be optimized to achieve a more uniform and efficient drying effect.

[0019] By comparing the first humidity evaluation feature and the second humidity evaluation feature with the internal humidity levels of different humidity ranges, the humidity level corresponding to the target mixer is determined.

[0020] Preferably, the system terminal analyzes and extracts the first and second humidity evaluation features to obtain drying rate features, humidity change sensitivity features, relative humidity fluctuation features, and humidity distribution features. The drying rate reflects the speed at which humidity decreases within the mixer and is a crucial basis for determining the humidity level. Humidity change sensitivity features help understand the mixer's sensitivity to humidity changes, further providing a reference for determining the humidity level. Relative humidity fluctuation features reflect the uniformity and stability of humidity within the mixer. Humidity distribution features help identify the existence of localized high-humidity areas or humidity gradients, providing an important basis for determining the humidity level. Subsequently, based on the analyzed features across multiple dimensions such as drying rate, humidity change sensitivity, relative humidity fluctuation, and humidity distribution, the system compares these features with predefined humidity level standards and determines the current humidity level of the target mixer according to the definitions and ranges in the standards. These standards are typically based on historical experience and industry standards and are used to classify humidity levels into different levels or ranges. The obtained humidity level comprehensively reflects the humidity status within the mixer, including humidity trends, stability, uniformity, and other aspects. The above process transforms complex humidity data into intuitive and understandable humidity levels, which not only helps in quickly understanding the humidity status of the mixer but also provides important reference for subsequent humidity control and optimization. By determining the humidity level, drying strategies can be formulated and adjusted more specifically, resulting in more efficient and precise mixer operation management.

[0021] Furthermore, this application provides a method for analyzing the change of the smallest humidity unit through the M monitoring time series data streams, setting a first humidity evaluation feature, and the method further includes: Based on the monitoring accuracy of the humidity sensor, determine the minimum humidity unit; Based on the M monitoring time series data streams, and using the smallest humidity unit as a reference, the humidity fluctuation frequency and humidity fluctuation amplitude are calculated. The first humidity evaluation feature is set by the humidity fluctuation frequency and humidity fluctuation amplitude.

[0022] Optionally, the system terminal determines a minimum humidity unit based on the monitoring accuracy of the humidity sensor. This minimum humidity unit forms the basis for subsequent humidity analysis and evaluation, ensuring the accuracy and reliability of the humidity data.

[0023] After obtaining the minimum humidity unit, the system terminal selects continuous data points from M monitoring time-series data streams. These data points cover a sufficient time range to reflect humidity fluctuations. Then, for each data point, it is determined whether it constitutes a complete humidity fluctuation cycle. A fluctuation cycle includes both increases and decreases in humidity. Based on the minimum humidity unit, it can be determined whether the change in humidity value reaches or exceeds this unit, thus defining a complete fluctuation cycle. Within the selected time range, the number of complete humidity fluctuation cycles is counted. This number is a direct indicator of the humidity fluctuation frequency. Next, the counted number of fluctuation cycles is divided by the selected time range to obtain the humidity fluctuation frequency per unit time. Then, within each complete humidity fluctuation cycle, the maximum and minimum humidity values ​​are found. These values ​​should be rounded down to the smallest humidity unit. Finally, the peak and trough values ​​of each fluctuation cycle are subtracted to obtain the humidity fluctuation amplitude for that cycle. Through the above process, the system terminal obtains the humidity fluctuation frequency and humidity fluctuation amplitude. Among them, humidity fluctuation frequency describes the number of times humidity changes within a certain period of time, reflecting how fast humidity changes; while humidity fluctuation amplitude describes the magnitude of humidity change, that is, the maximum difference in humidity value during the fluctuation process.

[0024] After obtaining the humidity fluctuation frequency and amplitude, the system terminal further sets the first humidity evaluation feature. This first humidity evaluation feature is a comprehensive description of the humidity change characteristics inside the mixer, including two key aspects: the frequency and amplitude of humidity changes. Through this feature, the dynamic changes in humidity inside the mixer can be understood more intuitively, providing important reference for subsequent humidity control and management. In summary, by determining the minimum humidity unit, calculating the humidity fluctuation frequency and amplitude, and setting the first humidity evaluation feature, a solid foundation is laid for comprehensively evaluating the humidity status inside the mixer.

[0025] Furthermore, this application provides a method for analyzing the humidity changes of the container's inner wall at the same time based on the M monitoring time series data streams, setting a second humidity evaluation feature, and the method further includes: Based on the M monitoring time series data streams, the data is split according to the time base to obtain N humidity collection data sets at time nodes, wherein each humidity collection data set includes humidity collection data at M key locations on the inner wall. Based on the humidity collection data set of the N time points, data change analysis is performed to obtain data change characteristics, which include the average humidity, maximum humidity, minimum humidity and humidity change rate within N-1 time periods. Based on the aforementioned data change characteristics, a second humidity evaluation feature is set.

[0026] Optionally, the system terminal determines a suitable time base based on actual needs and required analytical precision. This time base can be fixed, such as hourly, daily, or weekly data points, or it can be dynamic, set according to data fluctuations. Subsequently, the M monitoring time-series data streams are split according to the determined time base. This means cutting the continuous data stream into multiple time segments, each corresponding to a specific time node. During the splitting process, it is necessary to ensure that the data at each time node is complete, i.e., containing humidity data from all key locations on the mixer's inner wall. Then, for each time node, humidity data from the M key locations on the inner wall at that time are collected. These humidity data are then compiled into a dataset containing humidity information from all key locations at that time node. The above splitting and dataset formation process is then repeated for the entire time-series data stream until humidity data sets for N time nodes are obtained. These sets will cover the entire analysis period and provide a comprehensive humidity dataset.

[0027] After obtaining the humidity data set, the system terminal first clarifies the temporal relationship between N time points, dividing them into N-1 time periods. Each time period corresponds to the humidity data change between two adjacent time points. Then, for each time period, data change analysis is performed. Specifically, for all time points within that time period, the system terminal first extracts the humidity data at key locations on the corresponding inner wall. These data are then averaged to obtain the average humidity for that time period. This average reflects the overall humidity level within that time period. Next, within that time period, the humidity data at all time points are compared to identify the maximum and minimum humidity values. These two values ​​represent the highest and lowest humidity levels within that time period, reflecting the range of humidity fluctuation. Then, by comparing the humidity data at the start and end of the time period and calculating the difference between the humidity at the end and start, this difference is divided by the length of the time period to obtain the humidity change rate per unit time. The humidity change rate reflects how quickly humidity changes over time. Through the above calculation process, the system terminal obtains four data change characteristics for that time period: average humidity, maximum humidity, minimum humidity, and humidity change rate. Finally, repeat the above process to analyze each of the N-1 time periods and obtain the data change characteristics for each period. These characteristics will provide the system terminal with comprehensive information about the humidity changes inside the mixer, helping to better understand the trends and fluctuations in humidity, and providing a basis for subsequent humidity control and management.

[0028] After obtaining the data change characteristics, the system terminal sets a second humidity evaluation feature based on these calculated characteristics. This feature further describes the humidity change characteristics inside the mixer, combining the overall humidity level, fluctuations, and trends to provide the system terminal with a more comprehensive and in-depth humidity evaluation. In summary, this process involves splitting, analyzing changes, and extracting features from time-series data to ultimately obtain the second humidity evaluation feature, enabling a more accurate understanding of the humidity conditions inside the mixer.

[0029] If the degree of powder drying does not meet the preset powder drying threshold, efficiency compensation is performed based on the historical monitoring data of the humidity sensor, combined with the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, to determine the efficiency compensation parameters and the compensation reliability. In one embodiment, if the powder's dryness does not reach a preset drying threshold, the system terminal needs to adjust the drying process. In this case, the system terminal utilizes historical monitoring data from the humidity sensor, combined with the sensor's monitoring performance indicators and the mixer's operating performance indicators, to construct an efficiency compensation model. This model is then used to perform efficiency compensation, obtaining a compensation result that implicitly correlates efficiency compensation parameters with compensation reliability. The purpose of this is to identify key factors affecting drying efficiency and determine corresponding compensation parameters to improve drying efficiency. Simultaneously, the reliability of the obtained compensation parameters is evaluated to ensure the effectiveness and reliability of the adjustment measures. This step enables more accurate optimization of the powder drying process, ensuring product quality and production efficiency.

[0030] Furthermore, this application provides a method for efficiency compensation based on historical monitoring data from the humidity sensor, combined with the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, to determine efficiency compensation parameters and compensation reliability. The method further includes: Cross-correlation analysis is performed based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer to obtain cross-correlation indicators; Preferably, the system terminal first acquires the error of the humidity sensor, and then, through cross-correlation analysis, explores whether there is a correlation or mutual influence between the monitoring operating characteristic indicators, the operational operating characteristic indicators, and the humidity sensor error, obtaining a first cross-correlation index. Subsequently, the monitoring operating characteristic indicators, the operational operating characteristic indicators, and the mixer's operating status are combined, and another cross-correlation analysis is performed to obtain a second cross-correlation index. Afterwards, the first and second cross-correlation indices are fused for analysis to obtain the final cross-correlation index. Cross-correlation indices can quantitatively describe the degree of mutual influence between the humidity sensor and the mixer's operation. For example, does a change in the mixer's rotational speed cause fluctuations in the humidity value monitored by the humidity sensor, or does a change in the humidity monitored by the humidity sensor affect the mixer's operating efficiency? These cross-correlation indices not only help the system terminal better understand the relationship between humidity and mixer operation, but also provide important reference for subsequent humidity control and mixer optimization. In summary, cross-correlation analysis based on the monitoring operating characteristic indicators of the humidity sensor and the operational operating characteristic indicators of the target mixer can help the system terminal reveal the intrinsic connection between the two, providing strong support for optimized mixer operation and effective humidity control.

[0031] Based on the aforementioned cross-correlation index, an efficiency compensation model is established. This model is used to correct the implicit correlation between the error of the humidity sensor and the working state of the target mixer. Preferably, the system terminal establishes an efficiency compensation model based on cross-correlation indices to correct any implicit correlation between humidity sensor errors and the target mixer's operating state. Specifically, humidity sensors are susceptible to various factors in practical applications, such as environmental factors and equipment aging, leading to errors in the humidity data they monitor. Simultaneously, the mixer's operating state can also affect the humidity sensor's monitoring results; for example, changes in mixer speed and temperature may cause uneven humidity distribution or inaccurate sensor readings. The system terminal, through cross-correlation analysis, obtains cross-correlation indices that quantify the relationship between the humidity sensor and the mixer's operation, and then constructs the efficiency compensation model. Specifically, the system terminal first determines the model's inputs: the original measurement data from the humidity sensor, the target mixer's operating state data, and the cross-correlation indices. Then, it determines the model's outputs: the corrected efficiency compensation parameters and the compensation reliability. Subsequently, the model structure is determined based on the nature of the cross-correlation indices and the characteristics of the data. For example, if the relationship between the cross-correlation indices and the humidity sensor error is relatively simple and linear, a linear regression model is chosen; if the relationship is slightly more complex and not entirely nonlinear, a multinomial regression model is chosen. Specifically, taking a linear regression model as an example, the system terminal uses the humidity sensor error as the dependent variable and the mixer's operating status data and the original humidity sensor data as independent variables to construct a linear regression function. This function is then integrated into the linear regression model to build the initial efficiency compensation model. Next, cross-correlation indices are obtained and added as model constraints to the initial efficiency compensation model. These indices reflect the correlation between the humidity sensor error and the mixer's operating status, helping the model to more accurately predict and correct humidity data. Then, the initial efficiency compensation model is trained using a sample monitoring dataset. During training, the initial efficiency compensation model learns how to predict the humidity sensor error based on input features (including cross-correlation indices) and obtains efficiency compensation parameters and compensation confidence based on the prediction results. Finally, the performance of the initial efficiency compensation model is verified using a validation dataset. By comparing the predicted values ​​of the initial efficiency compensation model with the actual values, the model's correction effect on the humidity sensor error is evaluated. If the results are unsatisfactory, the initial efficiency compensation model should be optimized, for example, by adjusting the model parameters or reconstructing the linear regression function, until the predicted values ​​generated by the initial efficiency compensation model are not significantly different from the actual values. The trained initial efficiency compensation model should then be used as the final efficiency compensation model. The efficiency compensation model comprehensively considers the humidity sensor error and the mixer's operating status, correcting and compensating for the humidity sensor's monitoring data. This helps the system terminal more accurately obtain humidity information inside the mixer, reducing data deviations caused by sensor errors and changes in the mixer's operating status. This is of great significance for subsequent humidity control, mixer optimization, and product quality assurance.In summary, the efficiency compensation model based on cross-correlation indicators is a highly efficient tool that can correct the implicit correlation between humidity sensor errors and the operating status of the target mixer, improve the accuracy and reliability of humidity monitoring, and provide strong support for mixer operation optimization and product quality improvement.

[0032] Based on the efficiency compensation model, efficiency compensation is performed by combining the historical monitoring data of the humidity sensor to obtain implicit correlation compensation results, which include efficiency compensation parameters and compensation confidence.

[0033] Preferably, the system terminal utilizes a pre-built efficiency compensation model, combined with historical monitoring data from the humidity sensor, to perform efficiency compensation. Efficiency compensation refers to adjusting the humidity sensor data through the model to eliminate the implicit influence of the mixer's operating state on the humidity data. During the compensation process, the efficiency compensation model uses cross-correlation indicators to identify and correct errors in the humidity data based on historical data from the humidity sensor and the mixer's operating state data, generating implicit correlation compensation results. These implicit correlation compensation results include efficiency compensation parameters and compensation reliability. Efficiency compensation parameters are specific values ​​calculated by the efficiency compensation model during the compensation process, reflecting the degree of correlation between humidity sensor errors and the mixer's operating state. These parameters help understand the extent and manner in which the humidity sensor data is affected by the mixer's operating state. Compensation reliability assesses the reliability of the compensation results, based on the efficiency compensation model's fit to historical data and validation results, indicating the accuracy and reliability of the compensation results. The level of compensation reliability directly affects the degree of trust in the compensation results and their usage. In summary, after efficiency compensation based on the efficiency compensation model and historical monitoring data from the humidity sensor, implicit correlation compensation results, including efficiency compensation parameters and compensation reliability, can be obtained. These results will provide strong support for system terminals to more accurately understand the implicit correlation between humidity sensor data and the operating status of the mixer.

[0034] Furthermore, this application provides a method for obtaining cross-correlation indicators by performing cross-correlation analysis on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer. The method further includes: An error correction factor is set based on the error of the humidity sensor; Based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, a one-way cross-correlation analysis is performed using the error correction factor to obtain the first cross-correlation index. Optionally, based on the humidity sensor error, the system terminal first sets an error correction factor. This correction factor aims to quantify the measurement error of the humidity sensor for subsequent analysis and compensation. Then, an error correction component is randomly extracted from the error correction factor. This error correction component provides the system terminal with a direction for error correction. Next, the monitored operating characteristic indicators and the operational operating characteristic indicators are simultaneously correlated and paired. Then, a one-way cross-correlation analysis is performed based on the simultaneous correlation pairing results and the error correction component. The one-way cross-correlation analysis aims to identify any potential implicit correlations between the humidity sensor error and the target mixer's operating state. Through this analysis, a first cross-correlation index can be obtained, which quantifies the degree of correlation between the two, providing an important basis for subsequent efficiency compensation model construction and humidity data correction.

[0035] Based on the operating state of the target mixer, an operating state correction factor is set; Based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, a one-way cross-correlation analysis is performed using the operating state correction factor to obtain a second cross-correlation indicator. Optionally, based on the operating status of the target mixer, the system terminal sets an operating status correction factor. The main function of the operating status correction factor is to quantify the potential impact of differences in the mixer's operating status on the humidity sensor data, in order to more accurately analyze the correlation between the mixer's operating status and the humidity sensor data. Subsequently, using the aforementioned method of obtaining the first cross-correlation index, analysis is performed in conjunction with the monitoring operating characteristic index of the humidity sensor and the operating operating characteristic index of the target mixer. Then, using the set operating status correction factor, a one-way cross-correlation analysis is conducted on the monitoring operating characteristic index of the humidity sensor and the operating operating characteristic index of the mixer to explore in depth how changes in the mixer's operating status affect the monitoring data of the humidity sensor and to find the implicit correlation between the two. Through the one-way cross-correlation analysis, a second cross-correlation index can be obtained, which can quantify the degree of correlation between the mixer's operating status and the humidity sensor data.

[0036] Cross-correlation indicators are obtained by performing indicator fusion analysis using the first cross-correlation indicator and the second cross-correlation indicator.

[0037] Optionally, after obtaining the first and second cross-correlation indices, the system terminal performs index fusion analysis. Index fusion analysis organically combines the two indices to obtain a comprehensive cross-correlation index that reflects the degree of implicit correlation between the humidity sensor error and the target mixer's operating state. In the index fusion analysis, the system terminal first standardizes the first and second cross-correlation indices to eliminate biases caused by different units or numerical ranges, ensuring they have the same basis for comparison during the fusion process. Subsequently, based on historical experience, weights are assigned to the standardized first and second cross-correlation indices. The assigned weights represent the importance of each index in the comprehensive evaluation, reflecting their influence on the cross-correlation index. Once the weights are determined, the system terminal multiplies the values ​​of the first and second cross-correlation indices by their respective weights, adds the products, and divides by 2. In this way, the system terminal organically fuses the information of the first and second cross-correlation indices to obtain a comprehensive cross-correlation index that reflects the degree of implicit correlation between the humidity sensor error and the target mixer's operating state. This metric provides a more comprehensive and accurate perspective for the system terminal, enabling an understanding of the implicit relationship between the humidity sensor and the target mixer, and providing crucial support for the subsequent construction of efficiency compensation models and correction of humidity data.

[0038] Furthermore, this application provides monitoring performance indicators based on the humidity sensor and operational performance indicators of the target mixer, and performs unidirectional cross-correlation analysis using the error correction factor to obtain a first cross-correlation index. The method further includes: Randomly select the first error correction component from the error correction factors; The monitoring performance characteristics of the humidity sensor are used as independent variables, and the operating performance characteristics of the target mixer are used as dependent variables. They are then correlated and paired within the same time period to obtain the first one-way cross-correlation index. Based on the first error correction component and the first unidirectional cross-correlation index, a unidirectional cross-correlation analysis is performed to obtain the first cross-correlation index.

[0039] Optionally, the system terminal first analyzes the effect of each error correction factor. Different correction factors compensate for different types of errors or errors under specific conditions. Understanding the effects of these factors helps to more accurately select the appropriate correction component. Subsequently, based on an understanding of each correction factor, an error correction component is randomly selected as the first error correction component. This component will be used for subsequent cross-correlation analysis to help the system terminal more accurately understand the error situation of the humidity sensor.

[0040] After obtaining the first error correction component, the system terminal uses the monitoring performance indicators of the humidity sensor as the independent variable and the operating performance indicators of the target mixer as the dependent variable. For each time point, the monitoring performance indicator value is matched with the corresponding operating performance indicator value, forming a one-to-one pairing data. Then, the monitoring performance indicators in each pair are summed and divided by the number of pairs to obtain the mean of the monitoring performance indicators. Similarly, the mean of the operating performance indicators is calculated using the same method. Next, the difference between each pair of monitoring performance indicators and the operating performance indicator is calculated and then multiplied by the difference in the operating performance indicator. The products calculated for all pairs of pairs are summed and divided by the number of pairs to obtain the covariance of the paired data. Then, the Pearson correlation coefficient formula is introduced to calculate the first one-way cross-correlation index. This index reflects the linear correlation between the humidity sensor's monitoring data and the operating status of the target mixer; its value is between 0 and 1, with the absolute value closer to 1 indicating a stronger correlation.

[0041] After completing the correlation pairing, the system terminal evaluates the properties of the first error correction component, such as its magnitude, sign, and trend over time. This helps in understanding how the error correction process affects the relationship between variables. Similarly, the strength and direction of the first unidirectional cross-correlation index are analyzed. A higher value indicates a stronger unidirectional effect, while positive and negative values ​​represent positive and negative effects, respectively. Subsequently, a cross-correlation analysis is performed by combining the first error correction component and the first unidirectional cross-correlation index, observing whether their trends are consistent. If the change in the first error correction component is consistent with the trend of the first unidirectional cross-correlation index, this indicates a correlation between the error correction process and the unidirectional effect. Then, using the same method as calculating the first unidirectional cross-correlation index, the first cross-correlation index is calculated. This index quantifies the degree of correlation between the first error correction component and the first unidirectional cross-correlation index, providing the system terminal with a quantitative perspective for assessing the implicit correlation between humidity sensor error and the target mixer's operating state, offering important reference for subsequent efficiency and data correction work.

[0042] Based on the aforementioned data twin model of the mixer, the humidity level patterns under different operating conditions are simulated to construct a drying optimization space. The efficiency compensation parameters and compensation reliability are used as auxiliary variables for the optimization of the drying optimization space. In one embodiment, the system terminal sets different operating conditions according to actual needs, such as different material types, humidity requirements, and production speeds. Then, using a pre-constructed mixer data twin model, the operation of the mixer under these different conditions is simulated, with particular attention paid to the variation patterns of humidity levels. Subsequently, the simulation results of the mixer data twin model are analyzed to extract the trends and patterns of humidity level changes under different operating conditions, and key factors affecting humidity levels, such as material humidity, mixing time, and mixing speed, are identified. Based on the analysis results of humidity level patterns, the objectives and constraints for drying optimization are determined, and efficiency compensation parameters and compensation reliability are used as optimization auxiliary variables to construct a drying optimization space. The drying optimization space is used to explore how to achieve the best drying effect under various operating conditions. In this optimization space, efficiency compensation parameters and compensation reliability serve as important optimization auxiliary variables, guiding the system terminal on how to adjust drying parameters under different operating conditions to optimize the drying process. In this way, a more comprehensive understanding of the variation patterns of the drying process can be achieved, and the most suitable drying scheme for the current operating conditions can be found, thereby improving drying efficiency and quality.

[0043] In the drying optimization space, parameter optimization is performed to obtain the optimal combination of process parameters, and the powder drying control of the target mixer is optimized.

[0044] In one embodiment, within the constructed drying optimization space, the system terminal first defines the drying optimization objectives, such as achieving a specific humidity level, improving drying efficiency, and reducing energy consumption. Then, based on the simulation results of the mixer's data twin model, key parameters affecting the drying effect are determined, such as temperature, humidity, rotational speed, and material flow rate. Reasonable value ranges are then set for these key parameters based on historical experience to ensure the optimization process is conducted within the feasible domain. Next, the system terminal invokes the drying optimization space and performs iterative searches based on the set optimization objectives and parameter ranges. In each iteration, the drying optimization space generates a set of parameter combinations and uses the twin model for simulation evaluation, calculating indicators such as humidity level and drying efficiency under that combination. Then, based on the simulation evaluation results, the confidence value of each parameter combination is calculated to assess its merits. After multiple iterative searches, the drying optimization space converges to an optimal solution or a set of near-optimal solutions, i.e., the optimal process parameter combination. Once the optimal process parameter combination is determined, the system terminal applies the obtained optimal process parameters to the powder drying control of the target mixer, thereby optimizing the drying process. This not only improves drying efficiency and ensures product quality, but also reduces unnecessary energy consumption.

[0045] In summary, the embodiments of this application have at least the following technical effects: This application embodiment utilizes a humidity sensor to monitor the humidity of the inner wall of the target mixer container in real time, acquiring humidity data at key locations and establishing a data twin model of the mixer. By monitoring the trend of data changes, the humidity level of the mixer is evaluated, and the degree of powder drying is determined based on the humidity level, compared with a preset powder drying threshold. If the degree of drying does not meet the preset threshold, efficiency compensation is performed using historical monitoring data and operating characteristic indicators to determine compensation parameters and compensation reliability. Subsequently, based on the mixer data twin model, a drying optimization space is constructed, and parameter optimization is performed within this space to obtain the optimal combination of process parameters. In addition, by monitoring the time series data stream, the change of the smallest humidity unit is analyzed to set a first humidity evaluation feature. Simultaneously, the humidity change of the container inner wall is analyzed to set a second humidity evaluation feature. Finally, correction factors are set using the error of the humidity sensor and the operating status, and unidirectional cross-correlation analysis is performed to obtain cross-correlation indicators. Through the above steps, real-time monitoring, evaluation, compensation, and optimization of the powder drying process of the mixer are achieved, improving production efficiency and product quality. These technologies collectively address the technical problem of low powder drying quality stability caused by the inability to quickly debug and determine powder drying control optimization in powder drying scenarios where different material types or process conditions change. They enable real-time monitoring of humidity within the mixer, automatically adapting to different drying needs without the need for repeated testing and debugging for each type of material, thus improving the controllability of powder drying and ensuring the quality of powder drying.

[0046] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific and sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0047] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0048] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A real-time detection method for powder drying in a mixer based on a humidity sensor, characterized in that, The method includes: Based on a humidity sensor, the inner wall of the target mixer is monitored in real time to obtain real-time monitoring data. The real-time monitoring data includes humidity data collected at M key locations on the inner wall of the target mixer, where M is an integer greater than 0. The monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer are obtained, and a data twin model of the mixer is constructed. Based on the changing trend of the real-time monitoring data, the humidity level corresponding to the target mixer is evaluated, the degree of powder drying is determined according to the humidity level, and compared with the preset powder drying threshold. If the degree of powder drying does not meet the preset powder drying threshold, efficiency compensation is performed based on the historical monitoring data of the humidity sensor, combined with the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, to determine the efficiency compensation parameters and the compensation reliability. Based on the aforementioned data twin model of the mixer, the humidity level patterns under different operating conditions are simulated to construct a drying optimization space. The efficiency compensation parameters and compensation reliability are used as auxiliary variables for the optimization of the drying optimization space. In the drying optimization space, parameter optimization is performed to obtain the optimal combination of process parameters, and the powder drying control of the target mixer is optimized. Specifically, based on historical monitoring data from the humidity sensor, and combined with the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, efficiency compensation is performed to determine efficiency compensation parameters and compensation reliability, including: Cross-correlation analysis is performed based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer to obtain cross-correlation indicators; Based on the aforementioned cross-correlation index, an efficiency compensation model is established. This model is used to correct the implicit correlation between the error of the humidity sensor and the working state of the target mixer. Based on the efficiency compensation model, efficiency compensation is performed by combining the historical monitoring data of the humidity sensor to obtain implicit correlation compensation results, which include efficiency compensation parameters and compensation confidence.

2. The real-time detection method for powder drying in a mixer based on a humidity sensor as described in claim 1, characterized in that, Based on the changing trend of the real-time monitoring data, the humidity level corresponding to the target mixer is assessed, and the degree of powder dryness is determined according to the humidity level. The method includes: Based on the M key locations on the inner wall, M monitoring time series data streams are determined; By analyzing the changes in the smallest humidity unit through the M monitoring time series data streams, a first humidity evaluation feature is set, which includes a drying rate feature and a humidity change sensitivity feature. Based on the M monitoring time series data streams, the humidity change of the container inner wall at the same moment is analyzed, and a second humidity evaluation feature is set. The second humidity evaluation feature includes relative humidity fluctuation feature and humidity distribution feature. By comparing the first humidity evaluation feature and the second humidity evaluation feature with the internal humidity levels of different humidity ranges, the humidity level corresponding to the target mixer is determined.

3. The real-time detection method for powder drying in a mixer based on a humidity sensor as described in claim 2, characterized in that, The method involves analyzing the changes in the smallest humidity unit using the M monitoring time series data streams, and setting a first humidity evaluation feature. Based on the monitoring accuracy of the humidity sensor, determine the minimum humidity unit; Based on the M monitoring time series data streams, and using the smallest humidity unit as a reference, the humidity fluctuation frequency and humidity fluctuation amplitude are calculated. The first humidity evaluation feature is set by the humidity fluctuation frequency and humidity fluctuation amplitude.

4. The real-time detection method for powder drying in a mixer based on a humidity sensor as described in claim 2, characterized in that, Based on the M monitoring time series data streams, the method analyzes the humidity changes of the container's inner wall at the same time, and sets a second humidity evaluation feature. The method includes: Based on the M monitoring time series data streams, the data is split according to the time base to obtain N humidity collection data sets at time nodes, wherein each humidity collection data set includes humidity collection data at M key locations on the inner wall. Based on the humidity collection data set of the N time points, data change analysis is performed to obtain data change characteristics, which include the average humidity, maximum humidity, minimum humidity and humidity change rate within N-1 time periods. Based on the aforementioned data change characteristics, a second humidity evaluation feature is set.

5. The real-time detection method for powder drying in a mixer based on a humidity sensor as described in claim 1, characterized in that, Based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, a cross-correlation analysis is performed to obtain cross-correlation indicators. The method includes: An error correction factor is set based on the error of the humidity sensor; Based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, a one-way cross-correlation analysis is performed using the error correction factor to obtain the first cross-correlation index. Based on the operating state of the target mixer, an operating state correction factor is set; Based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, a one-way cross-correlation analysis is performed using the operating state correction factor to obtain a second cross-correlation indicator. Cross-correlation indicators are obtained by performing indicator fusion analysis using the first cross-correlation indicator and the second cross-correlation indicator.

6. The real-time detection method for powder drying in a mixer based on a humidity sensor as described in claim 5, wherein a first cross-correlation index is obtained by performing a one-way cross-correlation analysis based on the monitoring performance indicators of the humidity sensor and the operating performance indicators of the target mixer, using the error correction factor; the method includes: Randomly select the first error correction component from the error correction factors; The monitoring performance characteristics of the humidity sensor are used as independent variables, and the operating performance characteristics of the target mixer are used as dependent variables. They are then correlated and paired within the same time period to obtain the first one-way cross-correlation index. Based on the first error correction component and the first unidirectional cross-correlation index, a unidirectional cross-correlation analysis is performed to obtain the first cross-correlation index.

Images