A production detection method and system for high-insulation magnesium oxide powder

Abstract

The present invention relates to the technical field of production and detection of high-insulation magnesium oxide powder, and in particular to a production and detection method and system for high-insulation magnesium oxide powder, which are used to solve the problem that multiple performance indicators are difficult to detect in a coordinated manner and the process parameter control response is delayed. The system collects key performance indicators such as particle uniformity, insulation strength and material density in real time through a sensor array to construct operating status information. The priority coefficient of each indicator is calculated based on a dynamic weight allocation algorithm to divide the production line detection segment. A time series model is constructed for each segment and the ARIMA method is introduced to predict performance trends. Deviation evaluation indicators are generated in combination with an ideal process curve. Performance abnormality areas are identified based on the evaluation results, and a PID closed-loop control mechanism is constructed based on the leading indicators to adjust key process parameters in real time to achieve dynamic optimization and control of the target segmented area.

Description

Technical Field

[0001] The present invention relates to the technical field of material manufacturing and detection, and in particular to a production detection method and system for high-insulation magnesium oxide powder. Background Art

[0002] The production and testing technology for high-insulation magnesium oxide powder has evolved with increasing industrial demand. Its application in electrical equipment, cable filling materials, and high-temperature insulation materials places higher demands on the accuracy, efficiency, and automation of testing methods. However, existing testing technologies have limitations in terms of accuracy, real-time performance, and adaptability to complex operating conditions, and cannot fully meet the modern industrial demand for efficient and intelligent production of high-insulation magnesium oxide powder.

[0003] The existing technology has the following deficiencies:

[0004] At present, existing production detection technologies have shortcomings in terms of accurate evaluation of the insulation properties of high-insulation magnesium oxide powder, dynamic monitoring of particle uniformity, multi-parameter comprehensive detection and real-time feedback control, such as single detection dimension, high response delay and rough control strategy. These shortcomings lead to increased differences in product consistency, increased tolerance for production process deviations, and reduced production efficiency and material performance stability. Therefore, a production detection method and system for high-insulation magnesium oxide powder are proposed.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] This invention utilizes multi-dimensional testing of the high-insulation magnesium oxide powder production process, combined with real-time data acquisition and analysis techniques, to develop a systematic testing method and hardware architecture. Key performance indicators (such as insulation performance and particle uniformity) during the production process are dynamically monitored. A data processing module then classifies and optimizes the collected data to generate comprehensive evaluation results. If a performance indicator deviates from a preset range, the system automatically triggers a feedback control mechanism to adjust relevant process parameters, thereby achieving comprehensive optimization of the high-insulation magnesium oxide powder production process.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A production and testing method for high-insulation magnesium oxide powder comprises the following steps:

[0009] Step S1: Using a sensor array deployed at each key process node on the high-insulation magnesium oxide powder production line, performance indicators such as particle uniformity, insulation strength, and material density are collected as operating status information;

[0010] Step S2: Based on the operation status information, a dynamic weight allocation algorithm is used to calculate the priority coefficient of each performance indicator at each key process node;

[0011] Step S3: Divide the high-insulation magnesium oxide powder production line into test segments based on the priority coefficient, perform time series modeling on the performance indicators of each test segment, and compare the predicted trend with the ideal curve to form a comprehensive evaluation result;

[0012] Step S4: Based on the comprehensive evaluation results, the target segmented areas with performance deviations are identified, and the relevant process parameters are dynamically optimized and adjusted for the target segmented areas through the PID closed-loop control mechanism.

[0013] In a preferred embodiment, the operating status information of each key process node is collected, and the operating status information is three types of performance indicators, including particle uniformity, dielectric strength, and material density;

[0014] Collect particle size distribution information of high-insulation magnesium oxide powder particles and calculate particle uniformity;

[0015] The dielectric breakdown point voltage values ​​of a fixed number of magnesium oxide powder samples were measured, and the average value was taken as the insulation strength;

[0016] Calculate the density of a fixed number of magnesium oxide powder samples and take the average value as the material density.

[0017] In a preferred embodiment, a fixed time period is set forward from the current moment as the starting point as the current monitoring period;

[0018] Collect historical data of each performance indicator at each key process node within the current monitoring period and construct a historical data sequence;

[0019] Calculate the standard deviation of historical data in the historical data series as the degree of volatility.

[0020] In a preferred embodiment, a qualified threshold is defined for each performance indicator, and if the performance indicator is greater than or equal to the qualified threshold, the performance indicator is defined as qualified;

[0021] The number of performance indicators defined as qualified performance indicators in each key process node during the current monitoring period is counted and divided by the total number of performance indicators to obtain the historical qualified rate.

[0022] In a preferred embodiment, the difference between 1 and the historical qualified rate is multiplied by the degree of fluctuation to obtain a comprehensive weight value;

[0023] Divide the comprehensive weight value of each performance indicator by the sum of the comprehensive weight values ​​of all performance indicators on the key process node to obtain the priority coefficient;

[0024] Compare the values ​​of the priority coefficients of the three performance indicators corresponding to each key process node, and use the performance indicator corresponding to the maximum value as the dominant detection attribute of the node;

[0025] Critical process nodes with the same dominant detection attributes are divided into detection segments.

[0026] In a preferred embodiment, the values ​​of the performance indicators in the detection segments at different time points separated by the same time period in the current monitoring cycle are continuously collected to construct a stable time series;

[0027] Perform a stationary test on the time series. If the time series does not meet the stationary condition, perform difference processing on it and determine the difference order;

[0028] Based on the results of the autocorrelation function and partial autocorrelation function graph analysis, the ARIMA model order parameters (p, d, q) are set, where p represents the order of the autoregressive term, d represents the order of the difference term, and q represents the order of the moving average term.

[0029] After completing the setting of the ARIMA model order parameters, the least squares estimation method is used to solve, train and construct the ARIMA model.

[0030] In a preferred embodiment, the ARIMA model is applied to the time series within each detection segment, and a prediction sequence is generated by recursive calculation;

[0031] Construct an ideal curve sequence, calculate the difference between the predicted sequence and the ideal curve sequence at each time point, obtain the deviation value at each time point and take the average to obtain the deviation evaluation value;

[0032] Compare the deviation evaluation value with the preset evaluation threshold. If the deviation evaluation value is greater than or equal to the evaluation threshold, the comprehensive evaluation result is determined to be an abnormal evaluation result.

[0033] If the deviation evaluation value is less than the evaluation threshold, the comprehensive evaluation result is judged to be a normal evaluation result.

[0034] In a preferred embodiment, when the comprehensive evaluation result is determined to be an abnormal evaluation result, the detection segment is determined to have a performance deviation and is demarcated as a target segment area;

[0035] For the target segmented area, determine the corresponding adjustment parameters based on its dominant detection attributes;

[0036] The dominant detection attributes of the target segment area at the current moment are collected, compared with the ideal target value in the ideal curve sequence, the current error is calculated, and the adjustment value is generated based on the proportional, integral and differential control strategies.

[0037] In a preferred embodiment, the adjustment value is applied to the corresponding adjustment parameter in real time, quickly responding to deviations in the dominant detection attribute;

[0038] After each complete PID closed-loop control cycle is completed, the trend of the regulated time series is evaluated based on the ARIMA model, and the deviation evaluation value is recalculated;

[0039] If the deviation evaluation value is less than the evaluation threshold, it is judged that the control is effective, and the PID control loop can be exited to maintain the current process state;

[0040] Otherwise, the next control cycle will continue until the dominant performance indicator stabilizes and approaches the ideal curve, and the closed-loop control process ends.

[0041] A production and detection system for high-insulation magnesium oxide powder includes a data acquisition module, a data processing module, a segmented detection module, and an optimization control module. The functions of each module are as follows:

[0042] The data acquisition module collects performance indicators such as particle uniformity, insulation strength, and material density as operating status information through a sensor array deployed at key process nodes on the high-insulation magnesium oxide powder production line;

[0043] The data processing module uses the entropy weight method to calculate the priority coefficient of each performance indicator at each key process node based on the operating status information;

[0044] The segmented detection module divides the high-insulation magnesium oxide powder production line into detection segments based on priority coefficients, performs time series modeling on the performance indicators of each detection segment, and compares the predicted trend with the ideal curve to form a comprehensive evaluation result;

[0045] Based on the comprehensive evaluation results, the optimization control module identifies the target segmented areas with performance deviations and dynamically optimizes and adjusts the relevant process parameters in the target segmented areas through the PID closed-loop control mechanism.

[0046] The technical effects and advantages of the production and detection method and system of high-insulation magnesium oxide powder of the present invention are as follows:

[0047] The present invention realizes real-time collection of three types of key performance indicators, namely particle uniformity, insulation strength and material density, by using a sensor array to construct operating status information; based on the operating status information, a dynamic weight allocation algorithm is used to calculate the priority coefficient of each performance indicator to reflect the importance of each indicator at different process nodes; the production line is further divided into detection segments according to the priority coefficient, and a time series model is constructed for the segmented performance indicators. The trends are predicted by the ARIMA model and compared with the ideal process curve to form a deviation evaluation index; according to the deviation evaluation results, the target segmented area with performance deviation is identified, and a PID closed-loop control mechanism with the dominant performance indicator as the feedback object is constructed. The key process parameters are adjusted in real time to realize dynamic optimization and regulation of the target segmented area. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 Schematic diagram of the module structure of the high-insulation magnesium oxide powder production and detection system in an embodiment of the present invention.

[0049] Figure 2 Schematic diagram of the production line segment detection and feedback control process in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0051] Example 1, a production and detection method for high-insulation magnesium oxide powder, such as Figure 1 As shown, the following steps are included:

[0052] Step S1: Using a sensor array deployed at each key process node on the high-insulation magnesium oxide powder production line, performance indicators such as particle uniformity, insulation strength, and material density are collected as operating status information;

[0053] Step S2: Based on the operation status information, a dynamic weight allocation algorithm is used to calculate the priority coefficient of each performance indicator at each key process node;

[0054] Step S3: Divide the high-insulation magnesium oxide powder production line into test segments based on the priority coefficient, perform time series modeling on the performance indicators of each test segment, and compare the predicted trend with the ideal curve to form a comprehensive evaluation result;

[0055] Step S4: Based on the comprehensive evaluation results, the target segmented areas with performance deviations are identified, and the relevant process parameters are dynamically optimized and adjusted for the target segmented areas through the PID closed-loop control mechanism.

[0056] The specific implementation is as follows:

[0057] In step S1, a sensor array is deployed at each key process node of the high-insulation magnesium oxide powder production line to collect operating status information in real time. The operating status information includes three performance indicators: particle uniformity, insulation strength, and material density.

[0058] The particle size distribution information of high-insulation magnesium oxide powder particles is collected by laser particle size analyzer. After sampling, the particle size data is statistically processed to obtain the three characteristic particle size values ​​of D10, D50 and D90. The particle uniformity is calculated based on this parameter. This parameter reflects the concentration of the particle size distribution of high-insulation magnesium oxide powder particles. The specific calculation formula is as follows:

[0059] ;

[0060] Among them, U is the particle uniformity of each key process node. The smaller the value, the more uniform the particle distribution. D90 means that 90% of the particles are smaller than this particle size, D10 means that 10% of the particles are smaller than this particle size, and D50 means the median particle size.

[0061] The high-insulation magnesium oxide powder is divided into a fixed number of magnesium oxide powder samples. An increased voltage is applied to the magnesium oxide powder samples through a high-voltage breakdown voltage meter, and the dielectric breakdown point voltage values ​​of the magnesium oxide powder samples are recorded. To ensure data stability, a fixed number of magnesium oxide powder samples are measured in parallel, and the average of their dielectric breakdown point voltage values ​​is taken as the insulation strength of the key process node.

[0062] The volume and mass of the magnesium oxide powder samples were measured using a three-dimensional volume scanner and a high-precision electronic weighing instrument. The density of the magnesium oxide powder samples was calculated by dividing the mass by the volume. To ensure data stability, the density was calculated for a fixed number of magnesium oxide powder samples, and the average value was taken as the material density.

[0063] It should be noted that the setting of key process nodes is not fixed. In practical applications, the actual production links of high-insulation magnesium oxide powder shall prevail; the sensor array refers to an integrated device composed of multiple sensors of different types, which is installed at each key process node in the high-insulation magnesium oxide powder production line to collect parameters in the production process of high-insulation magnesium oxide powder. In this embodiment, it includes a laser particle size analyzer, a high-voltage breakdown voltage meter, a dielectric breakdown point, a three-dimensional volume scanner, and a high-precision electronic weighing instrument; the laser particle size analyzer is an analytical device for measuring the particle size distribution of powder based on the principle of laser scattering; the high-voltage breakdown voltage meter is a test equipment for measuring the breakdown voltage value of insulating materials under high electric fields; the dielectric breakdown point refers to the critical voltage value at which insulation failure occurs in a dielectric material at a certain thickness during the process of increasing voltage; the three-dimensional volume scanner is a measuring device that uses laser technology to extract the contours and reconstruct three-dimensionally irregular powder samples; the high-precision electronic weighing instrument is a mass measurement device with milligram-level resolution, which will not be described here.

[0064] In step S2, based on the three performance indicators of particle uniformity, insulation strength, and material density, a dynamic weight allocation algorithm is used to calculate the priority coefficient of each performance indicator at each key process node. The dynamic weight allocation algorithm determines the importance ranking of performance indicators in the high-insulation magnesium oxide powder production line based on the fluctuation degree of performance indicators and the historical qualified rate. The specific processing flow is as follows:

[0065] Starting from the current moment, a fixed time period is set forward as the current monitoring period. The historical data of each performance indicator at each key process node within the current monitoring period is collected, and a historical data sequence is constructed. The standard deviation of the historical data in the historical data sequence is calculated as the degree of fluctuation. The calculation formula for the degree of fluctuation is:

[0066] ;

[0067] in, is the degree of fluctuation, is the historical data in the historical data sequence, k is the index of the historical data, It represents the mean value of the performance indicator at the key process node, i is the index of the key process node, j is the index of the performance indicator, and n is the number of historical data in the historical data sequence. The larger the fluctuation degree, the stronger the volatility of the performance indicator, and the greater the instability of its impact on the production process.

[0068] A qualified threshold is defined for each performance indicator. If a performance indicator is greater than or equal to the qualified threshold, the performance indicator is defined as qualified. The number of performance indicators defined as qualified performance indicators in each key process node during the current monitoring period is counted and divided by the total number of performance indicators to obtain the historical qualified rate.

[0069] The fluctuation degree of the performance index and the historical pass rate are combined to calculate the comprehensive weight value. The calculation expression is as follows:

[0070] ;

[0071] in, is the comprehensive weight value, is the degree of fluctuation, The historical pass rate.

[0072] Normalize the comprehensive weight value of each performance indicator on each key process node, that is, divide the comprehensive weight value of each performance indicator by the sum of the comprehensive weight values ​​of all performance indicators on the key process node to obtain the priority coefficient. The value range of the priority coefficient is , and meet ,in, It is the priority coefficient of each performance indicator at each key process node.

[0073] It should be noted that the qualified threshold is a preset value obtained by professionals through experiments. In the embodiment, the dynamic weight allocation algorithm refers to an algorithm method that dynamically determines the relative importance of each indicator in the data processing or detection link based on the combined calculation of the fluctuation degree of each performance indicator at a specific process node and the historical qualified rate. The dynamic weight allocation algorithm can dynamically adjust the weight according to the actual data in the production process, avoiding the limitations brought by fixed weights, which will not be elaborated here.

[0074] In step S3, based on the priority coefficients of the three performance indicators of particle uniformity, insulation strength and material density at each key process node obtained in step S2, the production line is divided into sections for detection, and the numerical values ​​of the priority coefficients of the three performance indicators corresponding to each key process node are compared, and the performance indicator corresponding to the maximum value is used as the dominant detection attribute of the node.

[0075] Critical process nodes with the same dominant detection attributes are classified into the same detection segment. Specifically, the critical process node with the largest priority coefficient of particle uniformity constitutes the particle uniformity-dominated detection segment, the key normalization node with the largest priority coefficient of insulation strength constitutes the insulation strength-dominated detection segment, and the key process node with the largest priority coefficient of material density constitutes the material density-dominated detection segment.

[0076] Time series modeling is performed on the performance indicators in each detection segment. By continuously collecting the values ​​of the performance indicators in the detection segment at different time points separated by the same time period in the current monitoring cycle, a stable time series is constructed, and the time evolution trend is extracted using this time series.

[0077] Based on the time series, the autoregressive moving average integration model is used for modeling, and the time series is tested for stationarity. If the time series does not meet the stationary conditions, it is differentially processed to determine the differential order. Based on the results of the autocorrelation function and partial autocorrelation function image analysis, the ARIMA model order parameters (p, d, q) are set, where p represents the autoregressive order, d represents the differential order, and q represents the moving average order. After completing the setting of the ARIMA model order parameters, the least squares estimation method is used to solve, train and construct the ARIMA model.

[0078] After modeling is completed, the trained ARIMA model is applied to the time series in each detection segment, and the prediction sequence is generated through recursive calculation.

[0079] Construct an ideal curve sequence. The ideal curve sequence is given based on the timing characteristics or empirical process standards in high-quality production samples, representing the optimal performance indicator operation trajectory under the production state.

[0080] The difference between the predicted sequence and the ideal curve sequence is calculated point by point in time, the deviation value at each time point is obtained and the average is taken to obtain the deviation evaluation value. The larger the deviation evaluation value, the greater the deviation between the performance of the detection segment in the current monitoring cycle and the ideal operating state, reflecting that the process segment may have the risk of parameter drift or control failure; on the contrary, a smaller deviation evaluation value indicates that the detection segment is operating stably and the process control effect is good.

[0081] The deviation evaluation value is compared with the preset evaluation threshold. If the deviation evaluation value is greater than or equal to the evaluation threshold, the comprehensive evaluation result is determined to be an abnormal evaluation result. If the deviation evaluation value is less than the evaluation threshold, the comprehensive evaluation result is determined to be a normal evaluation result.

[0082] It should be noted that the evaluation threshold is a preset value obtained by professionals through experiments and will not be elaborated here.

[0083] In step S4, when the comprehensive evaluation result is determined to be an abnormal evaluation result, it is determined that the detection segment has a performance deviation and is demarcated as a target segment area.

[0084] For the target segmented area, the corresponding adjustment parameters are determined according to its dominant detection attribute, that is, a certain performance index among particle uniformity, insulation strength or material density.

[0085] If the dominant detection attribute of the target segmented area is particle uniformity, the adjustment parameters include but are not limited to the grinding motor speed, sieve aperture and feed rate;

[0086] If the dominant detection attribute of the target segmented area is dielectric strength, the adjustment parameters include calcination temperature, holding time and cooling rate;

[0087] If the dominant detection attribute of the target segmented area is material density, the adjustment parameters include compaction ratio, particle size composition and wet mixing ratio.

[0088] A PID closed-loop control mechanism is constructed with the dominant detection attribute as the control object and the adjustment parameter as the controlled variable. The PID closed-loop control mechanism is based on the following control logic:

[0089] Collect the dominant detection attributes of the target segment area at the current moment, compare them with the ideal target value in the ideal curve sequence, calculate the current error, and generate the adjustment value based on the proportional, integral and differential control strategies:

[0090] ;

[0091] in, is the adjustment value at the current time t, is the error value at the current time t, is the proportional coefficient, which adjusts the error response speed. is the integral coefficient, eliminating the steady-state error, is the differential coefficient, suppresses error fluctuation, T is the sampling cycle time, is the preset value, and all PID coefficients are preset by control system experts through step response experiments combined with the Ziegler-Nichols tuning method.

[0092] The above adjustment values ​​are applied to the corresponding adjustment parameters in real time, quickly responding to deviations in the dominant detection attributes. After adjustment, performance indicators are continuously collected, and error determination and control calculations are repeatedly performed to form a continuous closed-loop control.

[0093] To ensure control accuracy, after each complete PID closed-loop control cycle, the trend of the controlled time series will be evaluated again based on the ARIMA model, and the deviation evaluation value will be recalculated. If the deviation evaluation value is less than the evaluation threshold, the control is judged to be effective, the PID control loop can be exited, and the current process state can be maintained; otherwise, the next control cycle will be executed until the dominant performance indicator stabilizes and approaches the ideal curve, and the closed-loop control process ends.

[0094] Example 2, a production and detection system for high-insulation magnesium oxide powder, such as Figure 2 As shown, a production detection method for high-insulation magnesium oxide powder is implemented, including a data acquisition module, a data processing module, a segmented detection module, and an optimization control module. The modules are connected with electrical signals, and their functions are as follows:

[0095] The data acquisition module collects performance indicators such as particle uniformity, insulation strength, and material density as operating status information through a sensor array deployed at key process nodes on the high-insulation magnesium oxide powder production line;

[0096] The data processing module uses the entropy weight method to calculate the priority coefficient of each performance indicator at each key process node based on the operating status information;

[0097] The segmented detection module divides the high-insulation magnesium oxide powder production line into detection segments based on priority coefficients, performs time series modeling on the performance indicators of each detection segment, and compares the predicted trend with the ideal curve to form a comprehensive evaluation result;

[0098] Based on the comprehensive evaluation results, the optimization control module identifies the target segmented areas with performance deviations and dynamically optimizes and adjusts the relevant process parameters in the target segmented areas through the PID closed-loop control mechanism.

[0099] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0100] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0101] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0102] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device that includes the element.

[0103] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0104] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A production and detection method for high-insulation magnesium oxide powder, characterized in that: The following steps are involved: Step S1: Using a sensor array deployed at each key process node on the high-insulation magnesium oxide powder production line, performance indicators such as particle uniformity, insulation strength, and material density are collected as operating status information; Step S2: Based on the operation status information, a dynamic weight allocation algorithm is used to calculate the priority coefficient of each performance indicator at each key process node; Step S3: Divide the high-insulation magnesium oxide powder production line into test segments based on the priority coefficient, perform time series modeling on the performance indicators of each test segment, and compare the predicted trend with the ideal curve to form a comprehensive evaluation result; Step S4: Based on the comprehensive evaluation results, the target segmented areas with performance deviations are identified, and the relevant process parameters are dynamically optimized and adjusted for the target segmented areas through the PID closed-loop control mechanism.

2. The production and detection method of high-insulation magnesium oxide powder according to claim 1, characterized in that: Collecting operating status information of each key process node. The operating status information includes three types of performance indicators, including particle uniformity, dielectric strength, and material density; Collect particle size distribution information of high-insulation magnesium oxide powder particles and calculate particle uniformity; The dielectric breakdown point voltage values ​​of a fixed number of magnesium oxide powder samples were measured, and the average value was taken as the insulation strength; Calculate the density of a fixed number of magnesium oxide powder samples and take the average value as the material density.

3. The production and detection method of high-insulation magnesium oxide powder according to claim 2, characterized in that: Set a fixed time period from the current moment as the starting point as the current monitoring period; Collect historical data of each performance indicator at each key process node within the current monitoring period and construct a historical data sequence; Calculate the standard deviation of historical data in the historical data series as the degree of volatility.

4. The method for producing and detecting high-insulation magnesium oxide powder according to claim 2, wherein: A qualified threshold is defined for each performance indicator. If the performance indicator is greater than or equal to the qualified threshold, the performance indicator is defined as qualified. The number of performance indicators defined as qualified performance indicators in each key process node during the current monitoring period is counted and divided by the total number of performance indicators to obtain the historical qualified rate.

5. The method for producing and detecting high-insulation magnesium oxide powder according to claim 4, wherein: Subtract 1 from the historical pass rate and multiply it by the volatility to get the comprehensive weight value; Divide the comprehensive weight value of each performance indicator by the sum of the comprehensive weight values ​​of all performance indicators on the key process node to obtain the priority coefficient; Compare the values ​​of the priority coefficients of the three performance indicators corresponding to each key process node, and use the performance indicator corresponding to the maximum value as the dominant detection attribute of the node; Critical process nodes with the same dominant detection attributes are divided into detection segments.

6. The method for producing and detecting high-insulation magnesium oxide powder according to claim 5, wherein: Continuously collect the values ​​of performance indicators within the detection segment at different time points separated by the same time period within the current monitoring cycle to build a stable time series; Perform a stationary test on the time series. If the time series does not meet the stationary condition, perform difference processing on it and determine the difference order; Based on the results of the autocorrelation function and partial autocorrelation function graph analysis, the ARIMA model order parameters (p, d, q) are set, where p represents the order of the autoregressive term, d represents the order of the difference term, and q represents the order of the moving average term. After completing the setting of the ARIMA model order parameters, the least squares estimation method is used to solve, train and construct the ARIMA model.

7. The method for producing and detecting high-insulation magnesium oxide powder according to claim 6, wherein: Apply the ARIMA model to the time series within each detection segment and generate a forecast sequence through recursive calculation; Construct an ideal curve sequence, calculate the difference between the predicted sequence and the ideal curve sequence at each time point, obtain the deviation value at each time point and take the average to obtain the deviation evaluation value; Compare the deviation evaluation value with the preset evaluation threshold. If the deviation evaluation value is greater than or equal to the evaluation threshold, the comprehensive evaluation result is determined to be an abnormal evaluation result. If the deviation evaluation value is less than the evaluation threshold, the comprehensive evaluation result is judged to be a normal evaluation result.

8. The method for producing and detecting high-insulation magnesium oxide powder according to claim 5, wherein: When the comprehensive evaluation result is determined to be an abnormal evaluation result, the detection segment is determined to have a performance deviation and is demarcated as a target segment area; For the target segmented area, determine the corresponding adjustment parameters based on its dominant detection attributes; The dominant detection attributes of the target segment area at the current moment are collected, compared with the ideal target value in the ideal curve sequence, the current error is calculated, and the adjustment value is generated based on the proportional, integral and differential control strategies.

9. The method for producing and detecting high-insulation magnesium oxide powder according to claim 8, characterized in that: The adjustment value is applied to the corresponding adjustment parameter in real time, quickly responding to deviations in the dominant detection attribute; After each complete PID closed-loop control cycle is completed, the trend of the regulated time series is evaluated based on the ARIMA model, and the deviation evaluation value is recalculated; If the deviation evaluation value is less than the evaluation threshold, the control is judged to be effective, the PID control loop is exited, and the current process state is maintained; Otherwise, the next control cycle will continue until the dominant performance indicator stabilizes and approaches the ideal curve, and the closed-loop control process ends.

10. A production detection system for high-insulation magnesium oxide powder, based on the production detection method for high-insulation magnesium oxide powder according to any one of claims 1 to 9, characterized in that: It includes data acquisition module, data processing module, segment detection module and optimization control module. The functions of each module are as follows: The data acquisition module collects performance indicators such as particle uniformity, insulation strength, and material density as operating status information through a sensor array deployed at key process nodes on the high-insulation magnesium oxide powder production line; The data processing module uses the entropy weight method to calculate the priority coefficient of each performance indicator at each key process node based on the operating status information; The segmented detection module divides the high-insulation magnesium oxide powder production line into detection segments based on priority coefficients, performs time series modeling on the performance indicators of each detection segment, and compares the predicted trend with the ideal curve to form a comprehensive evaluation result; Based on the comprehensive evaluation results, the optimization control module identifies the target segmented areas with performance deviations and dynamically optimizes and adjusts the relevant process parameters in the target segmented areas through the PID closed-loop control mechanism.

Images