Product processing data acquisition method and system based on sensor

Abstract

The invention relates to the technical field of industrial data acquisition, in particular to a product processing data acquisition method and system based on a sensor, and the method comprises the steps: obtaining the operation parameters of stirring equipment at all moments in each fermentation process at a preset high sampling rate in the raw material fermentation of product processing, the operation parameters at least comprise output power and vibration data; calculating the fluctuation coefficient of each operation parameter in each fermentation process to obtain the coupling fluctuation degree of each fermentation process; calculating a time domain evaluation value of each fermentation process; calculating a frequency domain evaluation value of each fermentation process to obtain a comprehensive evaluation value representing the overall vibration abnormality of the equipment; a state discrimination value characterizing the instability of the operating state of the device is determined to select and execute different data storage policies. On the premise of ensuring that key data information is not lost, data storage and transmission loads are greatly reduced.

Description

Technical Field

[0001] This application relates to the field of industrial data acquisition technology, specifically to a sensor-based product processing data acquisition method and system. Background Technology

[0002] In product processing, monitoring the operating status of key equipment is crucial for ensuring process stability and product quality. By monitoring the operating status of key equipment, potential equipment failures can be identified and prevented in advance. However, traditional methods of collecting equipment operation data typically involve data collection at a fixed sampling frequency. Given the large number of devices requiring monitoring in a production environment, using a low sampling frequency may result in the loss of critical transient information during drastic changes or anomalies, leading to untimely or inaccurate identification of equipment failures. Conversely, using a high sampling frequency results in massive data volumes and high transmission and storage costs. Summary of the Invention

[0003] To address the aforementioned technical issues, a sensor-based method and system for acquiring product processing data are provided to resolve existing problems.

[0004] The solution to the technical problem in this application is to provide a sensor-based product processing data acquisition method and system, including the following steps: In a first aspect, embodiments of this application provide a sensor-based product processing data acquisition method, which includes the following steps: In the fermentation of raw materials for product processing, the operating parameters of the stirring equipment at each moment during each fermentation process are obtained with a preset high sampling rate. The operating parameters include at least the output power and vibration data. For each operating parameter, we analyze its fluctuation characteristics during a single fermentation process, calculate the fluctuation coefficient of each operating parameter during each fermentation process, and combine the synergy of the changes between different operating parameters to obtain the coupling fluctuation degree of each fermentation process. For each fermentation process, the time-domain evaluation value is calculated based on the distribution characteristics of the vibration data fluctuations in the time domain; the frequency-domain evaluation value is calculated based on the concentration of the energy distribution of the vibration data in the frequency domain; and the comprehensive evaluation value characterizing the overall vibration anomaly of the equipment is obtained by combining the time-domain evaluation value. For the current fermentation process, we analyze the differences between it and the adjacent fermentation processes in terms of vibration data trend changes, coupling fluctuation degree and comprehensive evaluation value, and determine the state discrimination value that characterizes the instability of equipment operation status, so as to select and implement different data storage strategies.

[0005] Preferably, the calculation of the fluctuation coefficient of each operating parameter during each fermentation process includes: Calculate the dispersion of each operating parameter at all times during each fermentation process; The differences between all adjacent time points under each operating parameter are positively fused to obtain the relative change of each operating parameter; Calculate the difference between the data of each operating parameter at the first time point and the data at the last time point, and use it as the overall trend. The volatility coefficient is the result of a positive fusion of the degree of dispersion, the relative change, and the overall trend.

[0006] Preferably, obtaining the coupling fluctuation degree for each fermentation process includes: Calculate the correlation between the output power and vibration data at all times during each fermentation process; The coupling volatility is positively correlated with the volatility coefficients and correlations of all operating parameters.

[0007] Preferably, the calculation of the time-domain evaluation value for each fermentation process includes: For each fermentation process, the extreme points of vibration data at all times are obtained, and the differences in vibration data corresponding to all two adjacent extreme points are positively fused as the fluctuation difference quantity. Mark the time corresponding to the maximum point as the significant time, count the time interval between each significant time and its adjacent significant time, and perform positive fusion on the difference between all two adjacent significant times as the interval difference quantity; The time-domain evaluation value is positively correlated with both the fluctuation difference and the interval difference.

[0008] Preferably, the calculation of the frequency domain evaluation value for each fermentation process includes: Frequency domain analysis was performed on the vibration data at all times during each fermentation process to obtain the spectrum. Calculate the passing frequency of the fan blades in the stirring shaft and define it as the fundamental frequency; mark the frequency components corresponding to all integer multiples of the fundamental frequency in the frequency spectrum as harmonic components; calculate the sum of the energies corresponding to all frequency components in the frequency spectrum as the total energy; The percentage of the sum of the energies of the fundamental frequency and all harmonic components in the total energy in the statistical frequency diagram is used as the frequency domain evaluation value for each fermentation process.

[0009] Preferably, the comprehensive evaluation value is positively correlated with both the time-domain evaluation value and the frequency-domain evaluation value.

[0010] Preferably, the calculation process of the state discrimination value is as follows: The difference in vibration data at all times between the current fermentation process and the previous fermentation process is calculated and denoted as the trend difference. Calculate the coupling fluctuation degree and the difference in the comprehensive evaluation value between the current fermentation process and the previous fermentation process, and record them as fluctuation difference and abnormal difference, respectively. The current state discrimination value of the fermentation process is the result of a positive fusion of trend difference, fluctuation difference and abnormal difference.

[0011] Preferably, the selection and execution of different data storage strategies includes: if the normalized state discrimination value is less than a preset threshold, then the sampling adjustment strategy is not executed; otherwise, the sampling adjustment strategy is executed.

[0012] Preferably, the high-resolution storage strategy refers to directly transmitting and storing the raw data of the current fermentation process; the low-resolution storage strategy refers to downsampling the raw data of the current fermentation process before transmitting and storing it.

[0013] Secondly, embodiments of this application also provide a sensor-based product processing data acquisition system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described sensor-based product processing data acquisition methods.

[0014] This application has at least the following beneficial effects: This application analyzes the fluctuation characteristics of various operating parameters and calculates the fluctuation coefficients of each operating parameter during each fermentation process. Its beneficial effect lies in quantifying the inherent fluctuation patterns of operating parameters from three dimensions: overall discreteness, instantaneous changes, and overall trends. It obtains the coupled fluctuation degree for each fermentation process, which is beneficial because it considers the coordinated change trends between operating parameters, effectively assessing synchronous fluctuations caused by process changes and asynchronous fluctuations caused by independent equipment failures. When process changes lead to a synchronous increase in power and vibration, the coupled fluctuation degree is amplified; when blade corrosion mainly causes abnormal vibration while power changes are small, the coupled fluctuation degree is suppressed, indicating that the stirring equipment experiences synchronous and severe fluctuations due to changes in process conditions. It calculates the time-domain evaluation value for each fermentation process, which is beneficial because it quantifies the intensity and instability of the vibration impact generated by the stirring blade during rotation in the time domain, thus initially capturing the periodic impact characteristics caused by corrosion. It calculates the frequency-domain evaluation value for each fermentation process, which is beneficial because it quantifies the concentration of vibration energy at the blade passage frequency and its harmonics in the frequency domain. This approach provides frequency domain evidence for periodic impacts, further capturing the characteristics of periodic impacts caused by corrosion. It yields a comprehensive evaluation value, which is beneficial because it comprehensively assesses potential local physical damage to the agitator blades from both time and frequency domains, indicating abnormal equipment vibration. It also determines the state discrimination value for the current fermentation process, selecting and implementing different data storage strategies. This is beneficial because by comparing the differences in vibration trends, coupling fluctuations, and comprehensive evaluation values ​​between adjacent fermentation processes, it can quantify the gradual changes in equipment state, thereby identifying early corrosion, scaling, or wear phenomena amidst signal aliasing. This significantly improves the accuracy of equipment operating state discrimination. When the equipment state is stable, data is downsampled and stored, greatly reducing the data volume to save storage space. When the equipment state is unstable, the raw data collected at a high sampling rate is directly stored to capture and retain higher-resolution fault characteristic information, ensuring that transient impact characteristics and subtle spectral components can be accurately captured. This significantly reduces the data storage and transmission load while ensuring that critical data information is not lost. Attached Figure Description

[0015] The sensor-based product processing data acquisition method of this application will be further described in detail below with reference to the accompanying drawings.

[0016] Figure 1 A flowchart illustrating the steps of a sensor-based product processing data acquisition method provided in this application embodiment; Figure 2 A flowchart illustrating the steps of the method for obtaining the coupling fluctuation degree of each fermentation process provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the sensor-based product processing data acquisition method and system proposed in this application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit the scope of this application.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0019] Please see Figure 1 The diagram illustrates a flowchart of a sensor-based product processing data acquisition method according to an embodiment of this application. The method includes the following steps: Step 1: In the fermentation of raw materials for product processing, the operating parameters of the stirring equipment at each moment during each fermentation process are obtained at a preset high sampling rate. The operating parameters include at least the output power and vibration data.

[0020] In food processing, raw material fermentation is a crucial step. During fermentation, the metabolic activities of microorganisms cause significant changes in the physical and chemical properties of the raw materials. By continuously stirring the raw materials, we can ensure the uniform mixing of materials in the fermentation tank, promote full contact between microorganisms and raw materials, and thus improve fermentation efficiency.

[0021] To ensure the efficiency of the fermentation process and the stability of product quality, it is necessary to monitor the operating data of the stirring equipment during fermentation. For example, voltage sensors, current sensors, and power sensors are deployed on the stirring motor inside the main equipment that stirs the raw materials to monitor the voltage, current, and output power of the stirring equipment during operation. Speed ​​sensors and vibration sensors are used to collect the speed of the stirring motor and the vibration data of the stirring shaft. Temperature sensors are used to collect the ambient temperature during the fermentation process. Among these, the operating data refers to all sensor data used to assess the health of the equipment.

[0022] In product processing scenarios such as stirring and fermentation, the early chronic corrosion caused by the long-term contact of stirring equipment with corrosive media results in weak abnormal signals induced by the performance degradation of the equipment. These signals are similar to normal fluctuation signals such as changes in material viscosity during fermentation, making it difficult to effectively identify the initial state of corrosion. Consequently, it is impossible to trigger an increase in the sampling frequency of the target equipment, resulting in a long-term low sampling frequency. This makes it impossible to capture key transient information in a timely manner, leading to untimely or inaccurate identification of equipment faults.

[0023] Based on the above analysis, during the fermentation process of the same batch of raw materials, the output power of the stirring motor and the vibration data of the stirring shaft at different times are collected at a preset high sampling rate, and the data are temporarily stored in a high-speed cache. The maximum-minimum normalization method is used to normalize the data in the high-speed cache. In this embodiment, all data are collected synchronously, and the high sampling rate of the data needs to satisfy the Nyquist theorem. The high sampling rate is set to 2kHz. As for other implementation methods, the implementer can set it according to the actual situation. The maximum and minimum value normalization method is a well-known technique and will not be described in detail here.

[0024] The output power and vibration data are defined as operating parameters, and the data of each operating parameter at each time point are obtained during each fermentation process. Thus, we have obtained the data of each operating parameter of the stirring equipment at each moment during each fermentation process.

[0025] Step 2: For each operating parameter, analyze its fluctuation characteristics during a single fermentation process, calculate the fluctuation coefficient of each operating parameter during each fermentation process, and combine the synergy of the changes between different operating parameters to obtain the coupling fluctuation degree of each fermentation process.

[0026] Furthermore, the flowchart of the method for obtaining the coupling fluctuation degree of each fermentation process provided in the embodiments of this application is as follows: Figure 2 As shown.

[0027] During the fermentation of raw materials, microorganisms secrete extracellular polysaccharides, while high molecular weight carbohydrates have thickening and gelling properties, which significantly increase the viscosity of the system. In order to ensure that fermentation can be fully carried out at different depths in the fermentation tank and to prevent the local viscosity of the raw materials from being too high, it is necessary to continuously stir the raw materials during the fermentation process.

[0028] During the fermentation of raw materials, the viscosity changes affect the stirring efficiency. To ensure thorough mixing, the stirring motor needs to output higher power to overcome the viscosity generated during fermentation. Simultaneously, the increased resistance of the high-viscosity fluid to the stirring blades leads to greater stress on the blades during rotation, potentially increasing vibration amplitude. Therefore, by analyzing the fluctuations in the operating parameters of the stirring motor, the fluctuation coefficient is calculated as follows: Calculate the dispersion of each operating parameter at all times during each fermentation process; In this embodiment, the degree of dispersion is measured by calculating the coefficient of variation of the data of each operating parameter at all times during each fermentation process. The coefficient of variation is a well-known technique and will not be elaborated here. As other implementation methods, implementers may use other methods of existing technology, such as variance, etc. This embodiment does not impose any special restrictions on this.

[0029] For each fermentation process, the difference in data for each operating parameter at two adjacent time points is calculated as the relative difference. The relative differences between all two adjacent time points under each operating parameter are positively fused to obtain the relative change of each operating parameter; In this embodiment, the calculation process of relative difference is as follows: calculate the average value of the data of each operating parameter at all times, calculate the absolute value of the difference between the data of each operating parameter at each time and the previous time, and take the ratio of the difference to the average value as the relative difference; secondly, the specific process of forward fusion is as follows: take the average value of the relative difference of each operating parameter at all two adjacent times as the relative change amount. In other implementation methods, the implementer can take the sum of the relative differences of each operating parameter at all two adjacent times as the relative change amount.

[0030] Calculate the difference between the data of each operating parameter at the first time point and the data at the last time point, and use it as the overall trend. In this embodiment, the absolute value of the difference between the data of each operating parameter at the first time point and the data at the last time point is calculated, and the ratio of this difference to the average value is positively mapped to the result, which is taken as the overall trend quantity. The positive mapping process is as follows: an exponential function is used for positive mapping, and the ratio of the difference to the average value is denoted as... ,Will The result is taken as the result of the positive mapping, where, It is an exponential function with the natural constant as its base; through the process of positive mapping, the result of the positive mapping is made non-negative.

[0031] By positively integrating the degree of dispersion, relative change, and overall trend, the fluctuation coefficient of each operating parameter during each fermentation process is obtained. In this embodiment, the specific process of positive fusion is as follows: the sum of the degree of dispersion, the relative change, and the overall trend is used as the fluctuation coefficient. In other implementation methods, the implementer may also use the product of the degree of dispersion, the relative change, and the overall trend as the fluctuation coefficient.

[0032] It should be noted that a greater degree of dispersion indicates that the operating parameter changes significantly during fermentation, and the data distribution is more dispersed, reflecting that the operating parameter is more unstable throughout the fermentation process. A greater relative difference indicates that the operating parameter changes significantly between two adjacent moments, and a greater relative change indicates that the operating parameter changes more frequently throughout the fermentation process. A greater overall trend indicates that the total change of the operating parameter from the start to the end of fermentation is significant. A larger fluctuation coefficient indicates that the operating parameter fluctuates violently during fermentation, and the stability of the fermentation process is poor, reflecting the possible existence of continuous unstable factors during fermentation, such as continuous changes in the viscosity of the raw materials.

[0033] Furthermore, during the stirring process, the changes in output power and vibration data are mainly affected by the viscosity changes caused by the fermentation of raw materials. By analyzing the consistency of the changing trends between output power and vibration data, and combining the fluctuation coefficient, the coupled fluctuation degree is determined, specifically: Calculate the correlation between the output power and vibration data at all times during each fermentation process; In this embodiment, the correlation is measured by calculating the reciprocal of the DTW distance. Specifically, the output power and vibration data are normalized separately. The reciprocal of the DTW distance between the normalized output power and normalized vibration data at all times during each fermentation process is taken as the correlation. The DTW distance is a well-known technique and will not be elaborated upon here. As for other implementations, the implementer can use other methods of the prior art, such as the Pearson correlation coefficient, etc. This embodiment does not impose any special restrictions on this. It should be noted that, to avoid the denominator being zero when calculating the ratio, a parameter tuning factor is added to the denominator. The value range of the parameter tuning factor is... In this embodiment, the parameter tuning factor is set to 1. In other implementation methods, the implementer can set it according to the actual situation.

[0034] The coupling volatility of each fermentation process is positively correlated with the volatility coefficients and correlations of all operating parameters; It should be noted that a positive correlation means that the dependent variable increases as the independent variable increases and decreases as the independent variable decreases.

[0035] In this embodiment, the product of the output power fluctuation coefficient, the vibration data fluctuation coefficient, and the correlation degree is calculated as the coupling fluctuation degree.

[0036] It should be noted that the greater the correlation, the more synergistic the trends of change between output power and vibration data; the greater the coupling fluctuation, the stronger the synchronous fluctuation experienced by the mixing equipment due to changes in process conditions. By considering the correlation, unrelated fluctuations caused by independent failures of the mixing equipment can be avoided.

[0037] Thus, the coupling fluctuation degree of each fermentation process is obtained.

[0038] Step 3: For each fermentation process, calculate the time-domain evaluation value of each fermentation process based on the distribution characteristics of the vibration data fluctuations in the time domain; calculate the frequency-domain evaluation value of each fermentation process based on the concentration of energy distribution of the vibration data in the frequency domain; combine the time-domain evaluation value to obtain a comprehensive evaluation value characterizing the abnormality of equipment vibration.

[0039] Furthermore, to ensure the fermentation effect of different batches of raw materials, each fermentation tank is used only for fermentation of a single raw material or a single product raw material, ensuring the specificity of microorganisms and the stability of the fermentation environment. Therefore, the stirring blades in a single fermentation tank will only be corroded by corrosive substances produced by the same fermentation bacteria.

[0040] During the stirring process of fermentation raw materials, the stirring blades are in long-term contact with corrosive substances such as organic acids produced after fermentation. These corrosive substances gradually erode the stirring blades, causing varying degrees of damage to the blade surface, which in turn leads to chronic corrosion, or even scaling or wear. As the degree of corrosion continues to deepen, the physical structure of the stirring blades will gradually change. For example, the shape of the blades may become distorted, and the surface roughness may increase. These changes will directly affect the stirring performance of the stirring blades, reducing stirring efficiency and increasing the resistance generated during the stirring process.

[0041] Secondly, the viscosity caused by raw material fermentation changes with fermentation time; the longer the fermentation time, the higher the viscosity, which has a greater impact on the operating status of the mixing equipment. Furthermore, corrosion of the mixing blades persists for a long time, and the degree of corrosion increases with prolonged contact, leading to more severe anomalies in the data after mixing. Therefore, when analyzing corrosion anomalies on the mixing blades of the mixing shaft, the impact of raw material fermentation on the mixing shaft needs to be considered.

[0042] Due to changes in the viscosity of the raw materials during fermentation, the vibration amplitude generated by the stirring blades changes accordingly. As viscosity increases, the overall vibration generated by the blades during stirring of the fermentation materials increases. Furthermore, due to corrosion of the blades, the vibration amplitude exhibits abrupt changes, which are particularly pronounced at certain locations. For example, when the blades rotate to a specific angle, the changes in vibration amplitude become similar. Based on this analysis, a time-domain evaluation value is calculated by examining the differences in vibration data over time during each fermentation process. Specifically: Obtain the extreme points of vibration data at all times during each fermentation process; In this embodiment, the AMPD (Automatic Multiscale-based Peak Detection) algorithm is used to obtain extreme points, including maximum and minimum points. The AMPD algorithm is a well-known technology and will not be described in detail here.

[0043] For each fermentation process, the differences in vibration data corresponding to all two adjacent extreme points are positively fused to form the fluctuation difference. In this embodiment, the absolute value of the difference between the vibration data corresponding to two adjacent extreme points is calculated, and the mean of the absolute values ​​corresponding to all two adjacent extreme points is used as the fluctuation difference.

[0044] Mark the time corresponding to the maximum point as the significant time, and count the time interval between each significant time and its adjacent significant time. In this embodiment, the time interval between each significant moment and its preceding significant moment is counted. It should be noted that the time interval is measured by counting the number of collections between two significant moments. Since each moment corresponds to a collection order, the time interval is taken as the total number of collections between the collection orders corresponding to two adjacent significant moments. However, the first significant moment is not calculated.

[0045] The differences in time intervals between all two adjacent significant moments are positively fused and used as the interval difference measure; In this embodiment, the mean of the absolute values ​​of the differences between the time intervals of all two adjacent significant moments is used as the interval difference.

[0046] The time-domain evaluation values ​​of each fermentation process were positively correlated with the fluctuation difference and the interval difference. In this embodiment, the product of the fluctuation difference and the interval difference is used as the time-domain evaluation value for each fermentation process. It should be noted that the larger the fluctuation difference, the greater the difference in vibration data between adjacent extreme points, reflecting significant fluctuations in the vibration of the stirring shaft during operation and more drastic changes in the resistance borne by the stirring blades; the larger the interval difference, the greater the difference in the time interval between two adjacent significant moments, reflecting more unstable motion state of the blades on the stirring shaft; the larger the obtained time-domain evaluation value, the more drastic the intensity of the vibration impact generated by the stirring blades during rotation, and the more unstable the time interval between occurrences, reflecting the significant instability of the impact intensity and period in the time domain of the vibration data, indicating more significant periodic impact characteristics caused by corrosion.

[0047] Secondly, corrosion of agitator blades is often non-uniform. Localized pitting or edge corrosion can lead to uneven mass distribution or changes in the hydrodynamic shape of the blades. When the blades rotate, these localized defects can trigger an additional stress concentration or fluid disturbance with each rotation. This disturbance manifests in vibration data as periodic impact components related to the blade frequency and its harmonics. Changes in viscosity also cause fluctuations in vibration data, which are related to the overall fluid resistance and thus appear as a wider frequency band in the spectrum.

[0048] Based on the above analysis, the frequency domain evaluation value is calculated by analyzing the variation characteristics of vibration data in the frequency domain, specifically as follows: Frequency domain analysis was performed on the vibration data at all times during each fermentation process to obtain the spectrum. In this embodiment, Fourier transform is used for frequency domain analysis to obtain the spectrum. The Fourier transform is a well-known technique and will not be described in detail here.

[0049] Calculate the passing frequency of the fan blades in the stirring shaft and define it as the fundamental frequency; It should be noted that the blade passage frequency, or blade frequency, is calculated using a well-known technique, and the specific calculation formula is as follows: Where N is the number of fan blades and RPM is the rotational speed of the stirring shaft.

[0050] The frequency components corresponding to all integer multiples of the fundamental frequency in the spectrum are marked as harmonic components; Calculate the sum of the energies corresponding to all frequency components in the frequency diagram, and use this sum as the total energy. The percentage of the sum of the energies of the fundamental frequency and all harmonic components in the total energy in the statistical frequency diagram is used as the frequency domain evaluation value for each fermentation process. It should be noted that the larger the frequency domain evaluation value, the more concentrated the vibration data energy is in the frequency domain on the frequency components that are strictly related to the rotation of the stirring blades. This indicates that the vibration energy of the blades is more likely to be periodic excitation, and more likely to be caused by local physical damage to the blades. It also indicates that the periodic impact characteristics caused by corrosion are more significant.

[0051] Furthermore, based on the time-domain evaluation value and the frequency-domain evaluation value, a comprehensive evaluation value is determined, specifically as follows: The comprehensive evaluation value of each fermentation process is positively correlated with both the time-domain evaluation value and the frequency-domain evaluation value; In this embodiment, the calculation process of the comprehensive evaluation value is as follows: A positive mapping is performed on the time-domain evaluation value, and its product with the frequency-domain evaluation value is used as the comprehensive evaluation value for each fermentation process; wherein, the positive mapping process is as follows: a logarithmic function is used for positive mapping, assuming the time-domain evaluation value is denoted as... Then The result is taken as the result of the positive mapping, where, The logarithmic function is based on 10. The time-domain evaluation result is smoothed through a positive mapping process.

[0052] It should be noted that the larger the comprehensive evaluation value, the more significant the fluctuations of the vibration data in the time and frequency domains during fermentation, indicating that the stirring blades have clear, severe and deteriorating local physical damage. This damage has produced strong abnormal characteristics in both the time and frequency domains, reflecting the greater instability of the equipment's operating status.

[0053] Thus, a comprehensive evaluation value for each fermentation process is obtained.

[0054] Step 4: For the current fermentation process, analyze the differences between it and the adjacent fermentation processes in terms of vibration data trend changes, coupling fluctuation degree and comprehensive evaluation value, determine the state discrimination value characterizing the instability of equipment operation state, and select and execute different data storage strategies.

[0055] Furthermore, when the fan blades react to changes in the viscosity of the fermentation feedstock, the operating parameters of the fan blades during the stirring process of the fermentation feedstock in adjacent fermentation processes show similarities, such as similar trends in vibration data. However, if the fan blades corrode, the differences in the trends of vibration data gradually increase with the degree of corrosion. Therefore, by analyzing the differences in vibration data between adjacent fermentation processes, as well as the differences in coupled fluctuation degree and comprehensive evaluation value, a state discrimination value is calculated, specifically: The difference in vibration data at all times between the current fermentation process and the previous fermentation process is calculated and denoted as the trend difference. In this embodiment, the DTW distance of vibration data at all times between the current fermentation process and the previous fermentation process is calculated and denoted as the trend difference. The DTW distance is a well-known technique and will not be described in detail here.

[0056] The difference in coupling volatility between the current fermentation process and the previous fermentation process is calculated and denoted as volatility difference. In this embodiment, the absolute value of the difference between the coupling fluctuation degree between the current fermentation process and the previous fermentation process is calculated and denoted as the fluctuation difference.

[0057] The difference between the overall evaluation value of the current fermentation process and the previous fermentation process is calculated and recorded as an abnormal difference. In this embodiment, the absolute value of the difference between the comprehensive evaluation value of the current fermentation process and the previous fermentation process is calculated and recorded as the abnormal difference.

[0058] By positively fusing trend differences, fluctuation differences, and abnormal differences, the state discrimination value of the current fermentation process is obtained; In this embodiment, the forward fusion process is as follows: using preset first weight, preset second weight, and preset third weight as weights, the trend difference, fluctuation difference, and abnormal difference are weighted and summed, and then normalized to serve as the state discrimination value of the current fermentation process. The sum of the preset first weight, preset second weight, and preset third weight is 1, the preset first weight is 0.3, the preset second weight is 0.3, and the preset third weight is 0.4. As for other implementation methods, the implementer can set them according to the actual situation. Next, the maximum and minimum value normalization method is used for normalization processing. The maximum and minimum value normalization method is a well-known technology and will not be described in detail here.

[0059] In this embodiment, the state discrimination value of the current fermentation process can be normalized based on the maximum and minimum state discrimination values ​​calculated during the historical fermentation process.

[0060] It should be noted that the greater the trend difference, the greater the difference in the trend of vibration data between the two fermentation processes; the greater the fluctuation difference, the greater the difference in the coupling fluctuation of operating parameters between the two fermentation processes; the greater the abnormal difference, the greater the significant difference in the operating state of the stirring shaft between the two fermentation processes, and the lower its stability; the larger the obtained state discrimination value, the greater the drastic change in the state of the stirring equipment between two adjacent fermentation processes, which most likely means that the fan blade corrosion or other mechanical damage is rapidly deteriorating and the equipment has entered an unstable state.

[0061] Furthermore, the larger the state discrimination value, the greater the risk of instability or abnormality in the operation of the mixing equipment. In order to monitor the operating status of the equipment more accurately, increasing the sampling frequency of the data during equipment operation can capture minute changes in the equipment operation process, which helps to detect potential abnormalities in a timely manner.

[0062] Therefore, based on the state discrimination value, a storage strategy for the operating data of the mixing equipment is selected, specifically as follows: If the normalized state discrimination value is greater than or equal to the preset threshold, the high-resolution storage strategy is executed; otherwise, the low-resolution storage strategy is executed. Specifically, the high-resolution storage strategy involves directly transmitting and storing the original data in the cache; the low-resolution storage strategy involves downsampling the original data in the cache before transmitting and storing it.

[0063] In this embodiment, the process of obtaining the preset threshold is as follows: based on the fermentation process under normal production batches in historical periods, the output power and vibration data of all fermentation processes are collected, and the state discrimination value of each fermentation process is obtained. The state discrimination value corresponding to the 95th percentile of the state discrimination values ​​of all fermentation processes is calculated as the preset threshold. In this embodiment, the preset threshold is 0.7. As for other implementation methods, the implementer can set it according to the actual situation. The 95th percentile is a well-known technology and will not be described in detail here. The downsampling processing method is a well-known technology and will not be described in detail here.

[0064] It should be noted that if the normalized state discrimination value is greater than or equal to the preset threshold, it indicates that the equipment state has changed significantly during this fermentation process. The stirring fan blades in the current raw material fermentation tank may have chronic corrosion, scaling, or early wear. In this case, the data collected at a high sampling frequency can reflect more detailed information. The original data with a high sampling rate is retained to improve the accuracy of subsequent anomaly analysis.

[0065] Based on the same inventive concept as the above methods, this application also provides a sensor-based product processing data acquisition system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described sensor-based product processing data acquisition methods.

[0066] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0067] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application. Therefore, any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application, without departing from the content of the technical solution of this application, shall fall within the protection scope of the technical solution of this application.

Claims

1. A sensor-based method for acquiring product processing data, characterized in that, The method includes the following steps: In the fermentation of raw materials for product processing, the operating parameters of the stirring equipment at each moment during each fermentation process are obtained with a preset high sampling rate. The operating parameters include at least the output power and vibration data. For each operating parameter, we analyze its fluctuation characteristics during a single fermentation process, calculate the fluctuation coefficient of each operating parameter during each fermentation process, and combine the synergy of the changes between different operating parameters to obtain the coupling fluctuation degree of each fermentation process. For each fermentation process, the time-domain evaluation value is calculated based on the distribution characteristics of the vibration data fluctuations in the time domain; the frequency-domain evaluation value is calculated based on the concentration of the energy distribution of the vibration data in the frequency domain; and the comprehensive evaluation value characterizing the overall vibration anomaly of the equipment is obtained by combining the time-domain evaluation value. For the current fermentation process, we analyze the differences between it and the adjacent fermentation processes in terms of vibration data trend changes, coupling fluctuation degree and comprehensive evaluation value, and determine the state discrimination value that characterizes the instability of equipment operation status, so as to select and implement different data storage strategies.

2. The sensor-based product processing data acquisition method as described in claim 1, characterized in that, The calculation of the fluctuation coefficients of each operating parameter during each fermentation process includes: Calculate the dispersion of each operating parameter at all times during each fermentation process; The differences between all adjacent time points under each operating parameter are positively fused to obtain the relative change of each operating parameter; Calculate the difference between the data of each operating parameter at the first time point and the data at the last time point, and use it as the overall trend. The volatility coefficient is the result of a positive fusion of the degree of dispersion, the relative change, and the overall trend.

3. The sensor-based product processing data acquisition method as described in claim 1, characterized in that, The process of obtaining the coupling fluctuation degree for each fermentation process includes: Calculate the correlation between the output power and vibration data at all times during each fermentation process; The coupling volatility is positively correlated with the volatility coefficients and correlations of all operating parameters.

4. The sensor-based product processing data acquisition method as described in claim 1, characterized in that, The calculation of the time-domain evaluation value for each fermentation process includes: For each fermentation process, the extreme points of vibration data at all times are obtained, and the differences in vibration data corresponding to all two adjacent extreme points are positively fused as the fluctuation difference quantity. Mark the time corresponding to the maximum point as the significant time, count the time interval between each significant time and its adjacent significant time, and perform positive fusion on the difference between all two adjacent significant times as the interval difference quantity; The time-domain evaluation value is positively correlated with both the fluctuation difference and the interval difference.

5. The sensor-based product processing data acquisition method as described in claim 1, characterized in that, The calculation of the frequency domain evaluation value for each fermentation process includes: Frequency domain analysis was performed on the vibration data at all times during each fermentation process to obtain the spectrum. Calculate the passing frequency of the fan blades in the stirring shaft and define it as the fundamental frequency; mark the frequency components corresponding to all integer multiples of the fundamental frequency in the frequency spectrum as harmonic components; calculate the sum of the energies corresponding to all frequency components in the frequency spectrum as the total energy; The percentage of the sum of the energies of the fundamental frequency and all harmonic components in the total energy in the statistical frequency diagram is used as the frequency domain evaluation value for each fermentation process.

6. The sensor-based product processing data acquisition method as described in claim 1, characterized in that, The comprehensive evaluation value is positively correlated with both the time-domain evaluation value and the frequency-domain evaluation value.

7. The sensor-based product processing data acquisition method as described in claim 1, characterized in that, The calculation process for the state discrimination value is as follows: The difference in vibration data at all times between the current fermentation process and the previous fermentation process is calculated and denoted as the trend difference. Calculate the coupling fluctuation degree and the difference in the comprehensive evaluation value between the current fermentation process and the previous fermentation process, and record them as fluctuation difference and abnormal difference, respectively. The current state discrimination value of the fermentation process is the result of a positive fusion of trend difference, fluctuation difference and abnormal difference.

8. The sensor-based product processing data acquisition method as described in claim 1, characterized in that, The selection and execution of different data storage strategies includes: if the normalized state discrimination value is greater than or equal to a preset threshold, then a high-resolution storage strategy is executed; otherwise, a low-resolution storage strategy is executed.

9. The sensor-based product processing data acquisition method as described in claim 8, characterized in that, High-resolution storage strategy refers to directly transmitting and storing the raw data of the current fermentation process; low-resolution storage strategy refers to downsampling the raw data of the current fermentation process before transmitting and storing it.

10. A sensor-based product processing data acquisition system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the sensor-based product processing data acquisition method as described in any one of claims 1-9.

Images