A rapeseed oil production line real-time monitoring method and system

Abstract

The present application relates to the technical field of image recognition, and particularly relates to a rapeseed oil production line real-time monitoring method and system, which acquires a sequence of oil cake time sequence images of a rapeseed oil production line in a running process; determines static forming parameters of oil cakes in multiple oil cake images in a preset period; determines dynamic forming parameters of the oil cakes in the preset period based on the static forming parameters corresponding to the multiple oil cake images in the preset period; determines a forming state of the oil cakes based on the dynamic forming parameters; and when the forming state is unformed, performs rapeseed oil production line maintenance alarm. The present application uses a double-layer feature system of static forming parameters and dynamic forming parameters to upgrade the discrimination of the oil cake forming state from single-frame instantaneous estimation to cumulative reliability evaluation in a continuous period, thereby reducing the misjudgment rate caused by instantaneous texture or light disturbance, and improving the accuracy of rapeseed oil production line anomaly detection.

Description

Technical Field

[0001] This invention relates to the field of image recognition technology, specifically to a method and system for real-time monitoring of rapeseed oil production lines. Background Technology

[0002] A rapeseed oil production line refers to a complete set of processing equipment and processes encompassing cleaning, pressing or leaching, refining, and packaging, from rapeseed raw materials to finished edible oil, achieving automated and continuous production. In a rapeseed oil production line, each process step must be closely integrated and operate collaboratively. Among these, the formation quality of the oil cake, as a crucial link between upstream and downstream processes, directly determines pressing efficiency and residual oil yield. Loose, broken, or unevenly thick oil cakes indicate problems in upstream processes such as improper rapeseed moisture control, deviations in roasting time and temperature, abnormalities in the machine's inner wall, uneven feeding, or drifting operating parameters. These problems will further lead to increased residual oil yield, decreased oil output, and even equipment blockage, severely impacting production continuity and oil quality. Therefore, real-time monitoring of the oil cake formation status is a core requirement for ensuring the efficient and high-quality operation of the rapeseed oil production line.

[0003] Existing problems: Current automated rapeseed oil production lines primarily rely on independent monitoring of each process stage, with overall control achieved through a multi-parameter collaborative judgment mechanism. Once an anomaly is detected, the fault is typically resolved by stopping the line for maintenance. This monitoring method, due to the fragmented parameters of each stage and the lack of direct online sensing of the oil cake formation status, fails to continuously and accurately identify end-to-end anomalies reflected in the oil cake's poor shape during production. This ultimately leads to frequent downtime for maintenance and significant resource waste. Summary of the Invention

[0004] This invention provides a method and system for real-time monitoring of rapeseed oil production lines to solve existing problems.

[0005] The present invention provides a method and system for real-time monitoring of a rapeseed oil production line, which adopts the following technical solution:

[0006] One embodiment of the present invention provides a method for real-time monitoring of a rapeseed oil production line. The method includes: acquiring a time-series image sequence of oil cake during the operation of the rapeseed oil production line; determining static forming parameters of the oil cake in multiple oil cake images within a preset time period; wherein the static forming parameters are used to characterize the probability of oil cake formation in a single oil cake image at the time of acquisition; determining dynamic forming parameters of the oil cake within the preset time period based on the static forming parameters corresponding to the multiple oil cake images within the preset time period; wherein the dynamic forming parameters are used to characterize the cumulative reliability of the oil cake maintaining a formed state within the preset time period; determining the forming state of the oil cake based on the dynamic forming parameters; and triggering a maintenance alarm for the rapeseed oil production line when the forming state is unformed.

[0007] Further, determining the static forming parameters of the oil cake in a single image of the oil cake includes: performing superpixel segmentation on the oil cake image to obtain multiple tile segmentation regions; extracting the area features and brightness features of each tile segmentation region, and determining the region static forming parameters of each tile segmentation region based on the area features and brightness features; and averaging the region static forming parameters of the multiple tile segmentation regions to obtain the static forming parameters of the oil cake in the oil cake image.

[0008] Further, determining the static forming parameters of a single tile segmentation region includes: determining a first local forming probability based on the area characteristics of the tile segmentation region; wherein the first local forming probability is used to characterize the local forming probability of the oil cake in the tile segmentation region determined based on the area characteristics; determining a second local forming probability based on the brightness characteristics of the tile segmentation region; wherein the second local forming probability is used to characterize the local forming probability of the oil cake in the tile segmentation region determined based on the brightness characteristics; and determining the area characteristics of all tile segmentation regions in the oil cake image where the tile segmentation region is located. The distribution of area features is used to determine a first weight value; wherein the first weight value is used to characterize the overall forming probability of the oil cake in the oil cake image based on the area feature distribution; a second weight value is determined based on the brightness feature distribution of all tile segmentation regions in the oil cake image where the tile segmentation region is located; wherein the second weight value is used to characterize the overall forming probability of the oil cake in the oil cake image based on the brightness feature distribution; based on the first weight value and the second weight value, the first local forming probability and the second local forming probability are weighted and summed to obtain the regional static forming parameter of the tile segmentation region.

[0009] Further, determining the dynamic forming parameters of the oil cake within the preset time period based on the static forming parameters corresponding to multiple oil cake images within the preset time period includes: determining a temporal difference index between two adjacent oil cake images based on the static forming parameters corresponding to multiple oil cake images within the preset time period; wherein the temporal difference index is used to characterize the degree of abrupt change in the static forming parameters between two oil cake images; clustering the temporal difference index to determine abnormal clusters; dividing the preset time period into multiple continuous segments based on the abnormal clusters; and determining the dynamic forming parameters of the oil cake within the preset time period based on the oil cake morphology characteristics of the oil cakes within the multiple continuous segments.

[0010] Furthermore, the step of clustering the time-series difference indicators to determine abnormal clusters includes: clustering multiple time-series difference indicators to obtain a first cluster and a second cluster; determining abnormal clusters and normal clusters in the first cluster and the second cluster; wherein, the cluster containing a smaller number of samples is the abnormal cluster, and the cluster containing a larger number of samples is the normal cluster.

[0011] Further, the step of dividing the preset time period into multiple continuous segments based on the abnormal clusters includes: taking the start time of the preset time period as the starting point, sequentially determining the cluster to which the time-series difference index belongs; when the time-series difference index belonging to the abnormal cluster is determined for the first time, taking the previous time of the time-series difference index as the end point of the current continuous segment; taking the next time of the time-series difference index as the starting point of the next continuous segment, determining the end point of the continuous segment; repeating the above process until the entire preset time period is traversed to obtain multiple continuous segments.

[0012] Further, determining the dynamic forming parameters of the oil cake within the preset time period based on the morphological features of the oil cake within multiple consecutive segments includes: for each consecutive segment, performing tile segmentation on the oil cake image at each time point within the consecutive segment to obtain multiple tile regions; for each tile region, calculating the tile similarity between the tile region and each tile region at a target time to obtain the morphological parameters of each tile region; wherein, the target time is used to characterize all subsequent times within the consecutive segment corresponding to the time point of the oil cake image where the tile region is located; the morphological parameters are the values ​​of the tile regions. The average similarity of the tiles; based on the similarity weighting, the morphological parameters of each tile region within the continuous segment are weighted and summed to obtain the dynamic forming parameters of the continuous segment; wherein, the similarity weighting is used to characterize the degree of continuous influence of the oil cake image where the tile region is located in the continuous segment; based on the oil cake morphological consistency weight, the dynamic forming parameters of each continuous segment are weighted and summed and averaged to determine the dynamic forming parameters of the oil cake within the preset time period; wherein, the morphological consistency weight is used to characterize the degree of consistency of oil cake morphology between two adjacent continuous segments.

[0013] Further, determining the forming state of the oil cake based on the dynamic forming parameters includes: normalizing the dynamic forming parameters of the oil cake within the preset time period; if the normalized dynamic forming parameters are greater than a preset forming threshold, the oil cake is determined to be in a formed state; if the normalized dynamic forming parameters are not greater than the preset forming threshold, the oil cake is determined to be in an unformed state.

[0014] Furthermore, after acquiring the time-series image sequence of rapeseed oil cake during the operation of the rapeseed oil production line, the method further includes: preprocessing the oil cake images in the time-series image sequence; wherein the preprocessing includes at least one of filtering, grayscale processing, and threshold segmentation processing.

[0015] Another embodiment of the present invention provides a real-time monitoring system for a rapeseed oil production line, comprising: an image acquisition device and a host computer, wherein:

[0016] The image acquisition device is installed at the cake discharge port of the rapeseed oil production line to acquire real-time images of the oil cake and send the images of the oil cake to the host computer.

[0017] The host computer is used to acquire a time-series image sequence of rapeseed cake during the operation of the rapeseed oil production line based on the oil cake images acquired by the image acquisition device; determine the static forming parameters of the oil cake in multiple oil cake images within a preset time period; wherein the static forming parameters are used to characterize the probability of oil cake formation in a single oil cake image at the time of acquisition; determine the dynamic forming parameters of the oil cake within the preset time period based on the static forming parameters corresponding to multiple oil cake images within the preset time period; wherein the dynamic forming parameters are used to characterize the cumulative reliability of the oil cake maintaining a formed state within the preset time period; determine the forming state of the oil cake based on the dynamic forming parameters; and trigger a maintenance alarm for the rapeseed oil production line when the forming state is unformed.

[0018] The beneficial effects of the technical solution of the present invention are:

[0019] In this embodiment of the invention, a time-series image sequence of rapeseed oil cake during the operation of the rapeseed oil production line is acquired; static forming parameters of the oil cake in multiple oil cake images within a preset time period are determined; wherein, the static forming parameters are used to characterize the probability of oil cake formation in a single oil cake image at the time of acquisition; based on the static forming parameters corresponding to multiple oil cake images within the preset time period, dynamic forming parameters of the oil cake within the preset time period are determined; wherein, the dynamic forming parameters are used to characterize the cumulative reliability of the oil cake maintaining a formed state within the preset time period; based on the dynamic forming parameters, the forming state of the oil cake is determined; when the forming state is unformed, a maintenance alarm for the rapeseed oil production line is triggered.

[0020] Thus, this invention upgrades the determination of the oil cake forming state from single-frame instantaneous estimation to cumulative reliability assessment over continuous time periods through a dual-layer feature system that combines static and dynamic forming parameters. This reduces the misjudgment rate caused by instantaneous texture or lighting disturbances, thereby improving the accuracy of anomaly detection in rapeseed oil production lines. On the other hand, by using superpixel segmentation to transform the oil cake image into tile segmentation regions, and constructing regional static forming parameters on two low-dimensional features of area and brightness, and then coupling them with global weights, the amount of data is compressed while retaining key morphological information, reducing the computational load and storage requirements of the host computer. On the one hand, based on the time-series difference index and the abnormal clusters obtained by clustering, the preset time period is divided into continuous segments, and a similarity weighting weight with decreasing time is introduced in each segment. This can effectively distinguish between short-term disturbances and continuous anomalies, avoid ineffective downtime, and thus improve the continuous running time of the production line. On the other hand, by comparing the normalized dynamic molding parameters with the preset molding threshold, the automatic determination of the molding state and the non-molded state is realized, and the maintenance alarm is triggered in real time when the molding is not formed. This shortens the response time from the occurrence of an anomaly to manual intervention and reduces the economic losses caused by risks such as increased residual oil rate and equipment blockage. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the real-time monitoring method for a rapeseed oil production line provided in an embodiment of the present invention;

[0023] Figure 2 This is a schematic diagram of the structure of the real-time monitoring system for a rapeseed oil production line provided in an embodiment of the present invention. Detailed Implementation

[0024] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a real-time monitoring method and system for a rapeseed oil production line proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0025] 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 invention pertains.

[0026] Before introducing the specific scheme of the above-mentioned real-time monitoring method and system for rapeseed oil production lines, we will first introduce the principles on which the above-mentioned real-time monitoring method and system for rapeseed oil production lines are based:

[0027] It should be noted that the rapeseed oil production line targeted in this embodiment of the invention adopts a general pressing process. The technical path is as follows: after cleaning and conditioning, the rapeseed undergoes oil separation through physical extrusion; under high pressure, the oil flows out from the oil outlet of the pressing chamber to form crude oil, while the remaining solids are pressed into solid flakes or block cakes and continuously discharged from the cake discharge port. These cakes are called oil cakes. The geometric morphology, structural integrity, and surface optical properties of the oil cake directly reflect the operating status of each upstream process step. Therefore, by conducting real-time morphological monitoring of the oil cake, the health status of the rapeseed oil production line can be non-invasively inferred.

[0028] Rapeseed cake can be clearly distinguished morphologically into two categories: "formed rapeseed cake" and "unformed rapeseed cake." Formed rapeseed cake, under continuous pressure, undergoes plastic deformation of the rapeseed grains, cell wall rupture, and oil release. The residual lipids between the grains form a continuous and stable gel-like porous structure. Macroscopically, it appears as large flakes or blocks with a dense surface and high hardness, exhibiting minimal deformation after leaving the pressing chamber. Unformed rapeseed cake, on the other hand, primarily undergoes elastic deformation. After pressure is released, the grains rebound and expand, the pore network is discontinuous and ineffective, macroscopically appearing as a powdery or small block structure with high oil content and a soft texture. It is prone to secondary fragmentation or collapse after leaving the pressing chamber.

[0029] From the perspective of oil quality, formed oil cakes have low residual oil content, resulting in lower losses during subsequent leaching or refining. Unformed oil cakes, due to the back-suction effect, have higher residual oil and impurity content, requiring increased chemical refining load. From the perspective of equipment reliability, formed oil cakes have an intact structure, reducing screw wear and minimizing the risk of conveyor blockage. Unformed oil cakes are loose and fragile, exacerbating wear on the screw and conveyor auger, and are prone to bridging blockages in areas such as the settling tank, forcing unplanned production line shutdowns.

[0030] The root causes of the dough sticks not taking shape can be summarized into the following four categories:

[0031] (1) Abnormalities in the seed roasting process: failure of temperature or moisture control leads to insufficient plasticity or excessive lubrication of rapeseed;

[0032] (2) Abnormal pressing process: The pressure curve is missing the low-pressure pre-pressing stage or the local pressure is insufficient, resulting in insufficient particle binding;

[0033] (3) Abnormal pressing chamber structure: The non-serrated smooth surface of the pressing screw and poor sealing of the pressing cage prevent the establishment of an effective pressure gradient;

[0034] (4) Abnormal raw material pretreatment: Incomplete cleaning introduces impurities such as sand and straw, which damage the particle bonding surface.

[0035] The real-time monitoring method and system for rapeseed oil production lines provided in this invention can achieve real-time monitoring of rapeseed oil production lines based on the above-mentioned mechanism. The specific solution of the real-time monitoring method and system for rapeseed oil production lines provided by this invention will be described in detail below with reference to the accompanying drawings.

[0036] Please see Figure 1 This illustrates a method for real-time monitoring of a rapeseed oil production line according to an embodiment of the present invention, comprising:

[0037] Step S110: Obtain the time sequence image of the rapeseed oil cake during the operation of the rapeseed oil production line.

[0038] It should be noted that the image acquisition device for collecting images of rapeseed cake can be positioned directly above the cake discharge port of the rapeseed oil production line. The image acquisition device can be an industrial-grade high-definition network camera with its lens optical axis perpendicular to the cake discharge direction to ensure the imaging plane is parallel to the surface of the rapeseed cake, thereby reducing perspective distortion. The camera outputs a real-time bitstream via a gigabit Ethernet interface at a frame rate of 1fps, meaning the time interval between two adjacent frames is 1 second. This sampling frequency is set based on the actual extrusion speed of the rapeseed cake: field measurements show that the extrusion speed of the rapeseed cake at the discharge port is not particularly fast, and a 1-second sampling period is sufficient to reduce data redundancy without missing key morphological information. The acquired original image sequence is denoted as:

[0039]

[0040] in, Indicates the time when the image was acquired.

[0041] Preferably, in one embodiment of the present invention, after step S110, the method may further include: preprocessing the oil cake images in the oil cake time-series image sequence; wherein the preprocessing includes at least one of filtering, grayscale processing and threshold segmentation processing.

[0042] It should be noted that the above filtering process can use a mean filter on the image. The main purpose of filtering is to eliminate Gaussian noise in the image. (This refers to the process of filtering the image.) The main purpose of grayscale conversion is to reduce the number of image channels and the amount of data, thereby simplifying subsequent image processing. The thresholding process described above can be performed using methods such as the Otsu thresholding method. The purpose of threshold segmentation is to ensure that only the oil cake information exists in the original image, thereby reducing interference from other device information and invalid calculations.

[0043] It should be further explained that the Otsu thresholding method described above is an adaptive threshold selection algorithm based on the image grayscale histogram. Its core idea is to traverse all possible grayscale thresholds to divide image pixels into foreground (target) and background classes, maximizing the inter-class variance and minimizing the intra-class variance. Ultimately, the grayscale level that maximizes the inter-class variance is determined as the optimal segmentation threshold, thus achieving adaptive binarization segmentation without manual parameter setting. It is understood that the Otsu thresholding method is a relatively mature and well-known technique in image processing technology; for its specific implementation, please refer to relevant technologies, which will not be elaborated upon in this embodiment.

[0044] Step S120: Determine the static forming parameters of the oil cakes in multiple oil cake images within a preset time period; wherein, the static forming parameters are used to characterize the oil cake forming probability of a single oil cake image at the time of its acquisition.

[0045] Preferably, in one embodiment of the present invention, step S120 may include: performing superpixel segmentation on the oil cake image to obtain multiple tile segmentation regions; extracting the area features and brightness features of each tile segmentation region respectively, and determining the regional static forming parameters of each tile segmentation region based on the area features and brightness features; and averaging the regional static forming parameters of the multiple tile segmentation regions to obtain the static forming parameters of the oil cake in the oil cake image.

[0046] Preferably, in one embodiment of the present invention, determining the regional static forming parameters of a single tile segmentation region includes: determining a first local forming probability based on the area characteristics of the tile segmentation region; wherein the first local forming probability is used to characterize the local forming probability of the oil cake in the tile segmentation region determined based on the area characteristics; determining a second local forming probability based on the brightness characteristics of the tile segmentation region; wherein the second local forming probability is used to characterize the local forming probability of the oil cake in the tile segmentation region determined based on the brightness characteristics; determining a first weight value based on the area characteristic distribution of all tile segmentation regions in the oil cake image where the tile segmentation region is located; wherein the first weight value is used to characterize the overall forming probability of the oil cake in the oil cake image determined based on the area characteristic distribution; determining a second weight value based on the brightness characteristic distribution of all tile segmentation regions in the oil cake image where the tile segmentation region is located; wherein the second weight value is used to characterize the overall forming probability of the oil cake in the oil cake image determined based on the brightness characteristic distribution; and weighting and summing the first local forming probability and the second local forming probability based on the first weight value and the second weight value to obtain the regional static forming parameters of the tile segmentation region.

[0047] It should be noted that the raw data for each moment's oil cake image includes a relatively short, fixed moment when rapeseed oil residue is extruded into oil cake from the production line; this can be understood as a static moment. From the above mechanistic analysis, it can be concluded that there is a static difference between an unformed and formed oil cake; therefore, this feature can be used to analyze the precipitation shape of the oil cake at a static moment. Taking the first... Taking a specific moment as an example, the details are as follows:

[0048] Step 1, with the first Using the original image at time step n as a basis, tile segmentation is performed using a superpixel segmentation algorithm to obtain the nth time step n. All tile segmentation areas at any given moment;

[0049] Step 2: After performing superpixel segmentation, the first pixel can be obtained. All of the moments Each tile divides the area;

[0050] With the first Taking a tile-divided area as an example ( ), and its corresponding regional static forming parameters The main steps to obtain it include:

[0051] (1) First, proceed with the first... All of the moments The area of ​​each tile segmentation region is obtained from the tile segmentation region, with the area of ​​the first tile segmentation region being the area of ​​the tile segmentation region. Taking the first region as an example, its corresponding area is the first region. The number of all pixels in a region is denoted as . ;

[0052] (2) Next, proceed with the first All of the regions The color of each tile segmentation region is obtained from the first tile segmentation region. Taking one region as an example, its corresponding color is the first. The average grayscale value of all pixels in a region is denoted as . ;

[0053] (3) Obtain the first Minimum area of ​​the tile segmentation region at time 1 and the The average area of ​​multiple tile-divided regions at a given time. ;

[0054] (4) Obtain the first The maximum brightness of all tile-divided regions at any given time. and the The average brightness of all tile segmentation regions at time 1 ;

[0055] (5) Use all the parameters that have been obtained to perform the first... Static forming parameters of the area divided by each tile The calculation formula is as follows:

[0056]

[0057] in, For normalization processing.

[0058] It should be further noted that the static forming parameters of the above-mentioned areas The calculation employs a weight-difference coupling method, mainly including:

[0059] (1) Global shaping probability weight calculation: the normalized value of the area dispersion of the entire image. Normalized value of brightness dispersion As a global formation probability weight and The smaller the value, the less formed the oil pancake becomes in the whole frame, and the weight value decreases accordingly, thus suppressing local features.

[0060] (2) Calculation of local forming probability: The first Area difference of each tile segmented area Consider the first local forming probability of the tile segmentation region, and the brightness difference value This is considered as the probability of second local formation.

[0061] (3) Perform coupling output.

[0062] It is understandable that: the current number The overall probability of the oil cake not forming at any given moment during extrusion is considered. During extrusion, formed oil cakes have a larger area and lower brightness compared to unformed ones. However, even in unformed extrusion, there are occasional instances where some areas have lower brightness and larger area. Therefore, it is necessary to use the overall state of the oil cake at any given moment as a weight to constrain this. The smaller, The smaller the value, the better. The greater the chance of the oil cake not forming properly when it is extruded at a certain moment, the more likely it is to fail. Then, by utilizing the first... The area of ​​the first tile region and the area of ​​the smallest region, and the area of ​​the first tile region The difference between the brightness of a tile area and the maximum brightness of all areas is used to determine the first... The probability that a region is an unformed region, the first The greater the area difference between the region with the smallest area and the region with the largest brightness, the greater the difference in brightness between the region with the largest brightness and the region with the largest area. This indicates that the... The higher the probability of a region being a formed region, the greater the likelihood of it being one. The purpose of weighting the global forming probability weights for the main body is to ensure that even if the main body has a large value, indicating it is a formed region, the probability of the first region being a formed region is still high. If the entire extrusion at time i results in an unformed region, then the weights are used to determine the region at time j. Restrictions are applied to certain areas, and vice versa.

[0063] The above method can be used to obtain the first The moment of the first Static forming parameters of the area divided by each tile , The larger it is, the more likely it is to be the first The more a tile divides an area, the more likely it is to form a shaped oil cake, and vice versa.

[0064] Step 3: Using the above method for the first... By acquiring the static forming parameters of all tile regions at a given time, the static parameters of each tile region can be obtained. Then, the static parameters of all tile regions are used to obtain the static parameters of the first tile region. Static forming parameters at a given time The specific calculation formula is as follows:

[0065]

[0066] The above scheme is based on the first All of the moments The mean value of the static forming parameters of the tile area represents the value of the first tile area. The static forming parameter of the overall shape of the oil cake extruded at a certain moment; the larger the value, the more significant the change. At a certain moment, there are many tile areas that exhibit the characteristics of large area and low brightness, i.e., the first... The higher the probability that the fried dough in the image at a given moment is a formed fried dough, the lower the probability is for the fried dough.

[0067] Thus, during the rapeseed oil production process on the production line, through the process of... The oil cake output at each moment was monitored and analyzed to obtain the data for the first time. The static forming parameters of the oil cake at a given moment.

[0068] It is understood that the superpixel segmentation algorithm described above is a computational method that clusters image pixels into locally homogeneous regions with similar characteristics such as color, brightness, or texture. Superpixel segmentation algorithms are relatively mature and well-known techniques in image processing technology. For specific implementation methods, please refer to related technologies; the embodiments of this invention will not be described in detail here.

[0069] Step S130: Based on the static forming parameters corresponding to multiple images of fried dough cakes within a preset time period, determine the dynamic forming parameters of the fried dough cakes within the preset time period; wherein, the dynamic forming parameters are used to characterize the cumulative reliability of the fried dough cakes maintaining their forming state within the preset time period.

[0070] Preferably, in one embodiment of the present invention, step S130 may include: determining a temporal difference index between two adjacent oil cake images based on the static forming parameters corresponding to multiple oil cake images within a preset time period; wherein the temporal difference index is used to characterize the degree of abrupt change in the static forming parameters between the two oil cake images; clustering the temporal difference index to determine abnormal clusters; dividing the preset time period into multiple continuous segments based on the abnormal clusters; and determining the dynamic forming parameters of the oil cake within the preset time period based on the oil cake morphology characteristics of the oil cakes within the multiple continuous segments.

[0071] Preferably, in one embodiment of the present invention, the above-mentioned clustering of time-series difference indicators to determine abnormal clusters includes: clustering multiple time-series difference indicators to obtain a first cluster and a second cluster; determining abnormal clusters and normal clusters in the first cluster and the second cluster; wherein, the cluster containing a smaller number of samples is an abnormal cluster, and the cluster containing a larger number of samples is a normal cluster.

[0072] Preferably, in one embodiment of the present invention, the above-mentioned method of dividing a preset time period into multiple continuous segments based on abnormal clusters includes: taking the start time of the preset time period as the starting point, sequentially determining the cluster to which the time series difference index belongs; when the time series difference index belonging to the abnormal cluster is determined for the first time, taking the previous time of the time series difference index as the end point of the current continuous segment; taking the next time of the time series difference index as the starting point of the next continuous segment, determining the end point of the continuous segment; repeating the above process until the entire preset time period is traversed to obtain multiple continuous segments.

[0073] Preferably, in one embodiment of the present invention, the above-mentioned determination of the dynamic forming parameters of the oil cake within a preset time period based on the morphological characteristics of the oil cake in multiple continuous segments includes: for each continuous segment, performing tile segmentation on the oil cake image at each moment within the continuous segment to obtain multiple tile regions; for each tile region, calculating the tile similarity between the tile region and each tile region at the target moment to obtain the morphological parameters of each tile region; wherein, the target moment is used to characterize all subsequent moments corresponding to the moment of the oil cake image located in the tile region within the continuous segment; the morphological parameter is the average tile similarity of the tile region; based on the similarity weighting weight, performing weighted summation on the morphological parameters of each tile region within the continuous segment to obtain the dynamic forming parameters of the continuous segment; wherein, the similarity weighting weight is used to characterize the degree of continuous influence of the oil cake image located in the tile region within the continuous segment; based on the oil cake morphological consistency weight, performing weighted summation and averaging on the dynamic forming parameters of each continuous segment to determine the dynamic forming parameters of the oil cake within the preset time period; wherein, the morphological consistency weight is used to characterize the degree of consistency of oil cake morphology between two adjacent continuous segments.

[0074] It should be noted that while the above method obtains the forming morphology characteristics of the oil cake at a static moment, using only the forming morphology characteristics of the oil cake at a static moment for monitoring the current rapeseed oil production line can easily fall into the short-time bias and one-sided analysis. The short-time bias specifically refers to the fact that during rapeseed oil production, the oil cake may only exhibit a loose form for a short period, which is not continuous and has little impact on overall production. One-sided analysis refers to the fact that rapeseed oil cake has its own texture; using static analysis methods may result in detecting this texture as oil cake blocks, leading to inaccurate judgments of rapeseed oil production line anomalies.

[0075] During continuous extrusion, the morphology of the oil cake exhibits significant temporal consistency differences: the formed oil cake, due to its dense gel porous structure, possesses high rigidity and undergoes minimal deformation under the action of subsequent material flow; while the unformed oil cake, due to weak interparticle bonding and loose structure, is prone to secondary fragmentation or collapse under continuous thrust. Therefore, the static forming parameters extracted from a single static frame may be misjudged due to transient texture interference. By introducing temporal consistency constraints, the static results can be dynamically corrected, thereby improving the accuracy and robustness of forming state discrimination.

[0076] Based on this logic, the changes in the oil cake image at continuous time points can be analyzed to obtain the dynamic forming parameters of the oil cake within a preset time period. (The last sentence appears to be incomplete and requires further context.) From the moment to the Taking a specific moment as an example, the steps for obtaining the dynamic forming parameters of the oil cake within a preset time period mainly include:

[0077] Step 1: First, obtain the static forming parameters of the oil cake using the method of obtaining the parameters of the oil cake. To the All at that moment Static forming parameters at a given moment.

[0078] Step two, then utilize the first To the All at that moment Obtain all static molding parameters at each moment. The temporal differences of static forming parameters at time step n, with the nth time step n as an example. Taking a specific moment as an example, its corresponding temporal difference index The calculation method is as follows:

[0079]

[0080] in, , and They represent the first , No. and the Static forming parameters of the oil cake at a given moment .

[0081] The above time series difference indicators The calculation logic is as follows: The temporal dissimilarity index at time step 1 describes the time series difference index at time step 2. The shape changes of the fried dough at a given moment in consecutive moments. The larger the value, the more it indicates the first... The shape of the oil cake at time 12 showed a significant difference from the shapes at consecutive time 12, that is, at time 12... The static forming parameters of the oil cake at a given moment may be inaccurate, and vice versa. The smaller the value, the more it indicates the number of... The more stable the shape of the fried dough sticks is at that moment.

[0082] By processing the static forming parameters of the oil cake at all times T' using the above method, the temporal difference index of the static forming parameters at all T' times can be obtained.

[0083] Step 3: Perform cluster analysis on the temporal differences of all static molding parameters to identify outlier clusters. The main steps include:

[0084] (1) With all The temporal difference indices of static forming parameters at each time point constitute a temporal difference index sequence, as follows: ;

[0085] (2) Use K-means clustering to cluster the time series difference index sequence. The number of clusters can be set to two. After clustering, two clusters can be obtained. The cluster with fewer samples in the cluster is regarded as an abnormal cluster, and the other cluster is regarded as a normal cluster.

[0086] The meaning of the above clustering is: at time... Within a given timeframe, the fried dough cakes are not necessarily formed or unformed at every moment; their shape can change. Therefore, by clustering using temporal differences across consecutive time points, if within all of these... If there is a morphological change at any given time, then there must be a corresponding temporal difference value that differs in magnitude from the other temporal difference values. This can be distinguished using the K-means clustering algorithm. If there is no morphological change in the oil cake during this process, then the difference between the abnormal cluster and the normal cluster will be significantly smaller after clustering.

[0087] Step 4: Obtain the dynamic forming parameters of oil cake extrusion using anomaly clusters and oil cake morphology characteristics over continuous time. The main steps include:

[0088] (1) First, the time-series difference sequence is divided using anomaly clusters to obtain continuous segments of the time-series difference sequence. The specific division method is as follows:

[0089] With sequence The first data The counting begins and continues until the preceding data point of a time-series difference value in an outlier cluster is encountered. This counted segment of time-series difference values ​​is then considered the first continuous segment. Next, starting from the time-series difference values ​​in the outlier cluster, the counting restarts until the preceding data point of a new time-series difference value in a new outlier cluster is encountered. Similarly, this counted segment of time-series difference values ​​is considered the new continuous segment. Using this method, the time-series difference sequence can be segmented to obtain... A series of consecutive segments.

[0090] (2) Next, the dynamic forming parameters of the oil cake extrusion in the current continuous segment are obtained by utilizing the morphological characteristics of the oil cake in each continuous segment, mainly including:

[0091] With the first Taking a continuous segment as an example, its corresponding dynamic morphological characteristics of the oil cake The acquisition method is as follows: First, for each time step in the continuous segment, the tile region in each time step is obtained using the tile segmentation algorithm; then, the tile region in each time step is obtained using the tile segmentation algorithm. The first in the consecutive segments Starting from the [time], detect the [time]th [time] respectively. The moment and the first The similarity of tile regions at all remaining time points in a consecutive segment (similarity can be obtained through a perceptual hashing algorithm) is used as the mean of all similarities as the morphological parameter at the first time point. Then, the morphological parameters at each time step are calculated using the method described above, thus obtaining the morphological parameters at each time step. It should be noted that only backward calculations are performed at each time step, starting with the first... Taking the first moment as an example, when calculating its morphological parameters, it is not related to the first moment. Similarity is calculated for moments prior to the first, only with the first... Similarity calculations are performed at times after a given time. ,in Indicates the first The total number of all moments in a continuous segment.

[0092] The above methods can be used to obtain Each shape parameter is then used to perform... The method for obtaining and calculating is as follows:

[0093]

[0094] in, Indicates the first The first in the consecutive segments The morphological parameters of the oil cake extruded at each moment.

[0095] The above-mentioned dynamic morphological characteristics of oil cake The calculation logic is as follows: using the morphological parameters at each time step within a continuous segment Based on this, an inverse time weight (i.e., similarity-weighted weight) is introduced. This design assigns greater weight to morphological parameters that appear earlier. The principle behind this weighting is as follows: if the morphology of earlier time points maintains high similarity, it indicates that the oil cake remains stable and undergoes minimal deformation during that period, thus requiring reinforcement; conversely, if the morphological similarity of earlier time points is low, it suggests that the oil cake has already deformed during continuous extrusion and will maintain this state in subsequent time points, thus requiring compensation. The weighted sum is obtained through the above weighted accumulation. The morphological stability of the entire continuous segment was quantified: The larger the value, the more stable the shape of the oil cake is and the less it changes during that period; conversely, the smaller the value, the more drastic the shape fluctuations.

[0096] By processing all the continuous segments using the above method, the dynamic morphological characteristics of the oil cake in each continuous segment can be obtained.

[0097] Then, the dynamic molding parameters are obtained, with the first... Taking a continuous segment as an example, its corresponding dynamic forming parameters The following is how to obtain it:

[0098] First, obtain the number The static shaping parameters at all times in the nth consecutive segment, with the nth segment as the first segment. Taking a specific moment as an example, the corresponding static forming parameters are: Secondly, utilizing the first Obtain all possible static shaping parameters in the nth consecutive segment. The mean of static shaping parameters at all times in a continuous segment Finally utilize and conduct The calculation method is as follows:

[0099]

[0100] The calculation logic of the above calculation method is as follows: The larger the value, the more it indicates the number of... The larger moments in a continuous segment are more likely to be formed oil cakes under static observation; The larger the value, the more it indicates the number of... The shape of the oil cake at each moment in a continuous segment is relatively stable in subsequent moments. Therefore, the larger the two values ​​are, the more stable the shape of the oil cake at each moment in the segment. The greater the likelihood of the oil cake forming in a continuous segment, the better its continuity.

[0101] Finally, through all The dynamic shaping parameters for the current preset time period are obtained from each continuous segment. The specific method of obtaining it is as follows:

[0102]

[0103] in, Indicates the first The first in the consecutive segments Static forming parameters at a given moment; Indicates the first Dynamic forming parameters for a continuous segment.

[0104] The calculation logic of the above formula is as follows: The dynamic forming parameters of each continuous segment are weighted and averaged. Taking the first segment as an example... Taking a continuous segment as an example, the consistency weight can be expressed as: Its purpose is to determine the first The consecutive segments and the first Are the shapes of the fried dough cakes consistent in each consecutive segment? The first moment in a consecutive segment is data from the outlier cluster, i.e., it is related to the first moment. The oil cake exhibits static morphological differences at different times within a series of consecutive segments, therefore, the first... The static shaping parameters of the first moment in the consecutive segments are used to compare with the... The average values ​​of all static forming parameters in each consecutive segment are compared. The closer the ratio is to 1, the better the formation of the first segment. The shape of all moments in the consecutive segments and the first segment The closer the forms are at each moment, the higher the consistency weight. The larger the consistency weight, the greater the dynamic shaping parameters for the current time period. , The larger the value, the higher the number of... To the Within a given time period, the likelihood of the fried dough sticks forming a shaped dough stick is higher, and vice versa.

[0105] It should be noted that since the last continuous segment is the current time, weight calculation and comparison cannot be performed, so the default weight is 1.

[0106] The above method can be used to obtain the dynamic molding parameters of rapeseed oil production line at different time periods during the production process.

[0107] At this point, the dynamic forming parameters of the rapeseed oil cake in the production process have been obtained.

[0108] It should be further noted that: the K-means clustering algorithm described above is an unsupervised iterative optimization algorithm. The tile segmentation algorithm described above refers to a computational method for dividing an image of a pancake into several local homogeneous regions (tiles), such as superpixel segmentation algorithms. The perceptual hashing algorithm described above is an algorithm that maps image content into compact binary or integer fingerprints, commonly used for image deduplication, copyright detection, and similarity assessment. The K-means clustering, tile segmentation, and perceptual hashing algorithms described above are all relatively mature and well-known technologies. For their specific implementation methods, please refer to relevant technologies; the embodiments of this invention will not elaborate further.

[0109] Step S140: Determine the forming state of the oil cake based on dynamic forming parameters.

[0110] Preferably, in one embodiment of the present invention, step S140 may include: normalizing the dynamic forming parameters of the oil cake within a preset time period; if the normalized dynamic forming parameters are greater than a preset forming threshold, the oil cake is determined to be in a formed state; if the normalized dynamic forming parameters are not greater than the preset forming threshold, the oil cake is determined to be in an unformed state.

[0111] Step S150: When the forming state is not formed, trigger the rapeseed oil production line maintenance alarm.

[0112] It should be noted that after obtaining the static and dynamic forming parameters of the rapeseed oil cake during the rapeseed oil production process, the rapeseed oil production line can be monitored in real time using the dynamic forming parameters. The main steps include:

[0113] Step 1: Determine the morphology of the rapeseed oil cake by analyzing the dynamic forming parameters during the rapeseed oil production process within a preset time period. The specific determination method is as follows: First, normalize the dynamic forming parameters of the rapeseed oil extrusion cake within the preset time period; then, establish a parameter-corresponding oil cake morphology model based on experience. Table 1 shows the experience model for a certain application scenario:

[0114] Table 1 Empirical Model

[0115]

[0116] Then, the threshold method is used for judgment. In this embodiment of the invention, the forming threshold is 0.85. When the dynamic forming parameter of the oil cake is less than 0.85 within a preset time period, the oil cake is considered not to be formed.

[0117] For the formed oil cake, it is determined that the rapeseed moisture content, roasting time and temperature, inner wall of the machine, feeding uniformity, and operating parameters in the rapeseed oil production line are normal.

[0118] For unformed oil cakes, it is believed that there are abnormalities in the rapeseed moisture content, roasting time and temperature, inner wall of the machine, uniformity of feeding, and operating parameters in the rapeseed oil production line, and maintenance of the rapeseed oil production line is required.

[0119] Please see Figure 2 This illustrates another embodiment of the present invention, a real-time monitoring system 200 for a rapeseed oil production line, comprising: an image acquisition device 210 and a host computer 220, wherein:

[0120] Image acquisition device 210 is installed at the cake discharge port of rapeseed oil production line to acquire real-time images of oil cake and send the oil cake images to host computer 220.

[0121] The host computer 220 is used to acquire a time-series image sequence of rapeseed cake during the operation of the rapeseed oil production line based on the oil cake images acquired by the image acquisition device; determine the static forming parameters of the oil cake in multiple oil cake images within a preset time period; wherein, the static forming parameters are used to characterize the probability of oil cake forming in a single oil cake image at the time of acquisition; based on the static forming parameters corresponding to multiple oil cake images within the preset time period, determine the dynamic forming parameters of the oil cake within the preset time period; wherein, the dynamic forming parameters are used to characterize the cumulative reliability of the oil cake maintaining the forming state within the preset time period; based on the dynamic forming parameters, determine the forming state of the oil cake; and when the forming state is unformed, trigger a maintenance alarm for the rapeseed oil production line.

[0122] This invention is now complete.

[0123] In summary, in this embodiment of the invention, a time-series image sequence of rapeseed oil cake during the operation of the production line is obtained; static forming parameters of the oil cake in multiple oil cake images within a preset time period are determined; wherein, the static forming parameters are used to characterize the probability of oil cake formation in a single oil cake image at the time of acquisition; based on the static forming parameters corresponding to multiple oil cake images within the preset time period, dynamic forming parameters of the oil cake within the preset time period are determined; wherein, the dynamic forming parameters are used to characterize the cumulative reliability of the oil cake maintaining a formed state within the preset time period; based on the dynamic forming parameters, the forming state of the oil cake is determined; and when the forming state is unformed, a maintenance alarm for the rapeseed oil production line is triggered. This invention utilizes a dual-layer feature system combining static and dynamic forming parameters to upgrade the determination of oil cake forming status from single-frame instantaneous estimation to cumulative reliability assessment over continuous time periods. This reduces the misjudgment rate caused by instantaneous texture or lighting disturbances, thereby improving the accuracy of anomaly detection in rapeseed oil production lines. Furthermore, by using superpixel segmentation to convert the oil cake image into tile-segmented regions and constructing regional static forming parameters based on two low-dimensional features: area and brightness, coupled with global weights, this process compresses the data volume while preserving key morphological information, reducing the computational load and storage requirements of the host computer. On the one hand, based on the time-series difference index and the abnormal clusters obtained by clustering, the preset time period is divided into continuous segments, and a similarity weighting weight with decreasing time is introduced in each segment. This can effectively distinguish between short-term disturbances and continuous anomalies, avoid ineffective downtime, and thus improve the continuous running time of the production line. On the other hand, by comparing the normalized dynamic molding parameters with the preset molding threshold, the automatic determination of the molding state and the non-molded state is realized, and the maintenance alarm is triggered in real time when the molding is not formed. This shortens the response time from the occurrence of an anomaly to manual intervention and reduces the economic losses caused by risks such as increased residual oil rate and equipment blockage.

[0124] 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 principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for real-time monitoring of a rapeseed oil production line, characterized in that, The method includes: Obtain a time-series image sequence of rapeseed oil cake during the operation of the rapeseed oil production line; Determine the static forming parameters of oil cakes in multiple oil cake images within a preset time period; wherein, the static forming parameters are used to characterize the probability of oil cake formation in a single oil cake image at the time of its acquisition; Based on the static forming parameters corresponding to multiple images of the fried dough within the preset time period, the dynamic forming parameters of the fried dough within the preset time period are determined; wherein, the dynamic forming parameters are used to characterize the cumulative reliability of the fried dough maintaining its forming state within the preset time period; Based on the dynamic forming parameters, the forming state of the oil cake is determined; When the forming state is unformed, an alarm for maintenance of the rapeseed oil production line is triggered.

2. The method for real-time monitoring of a rapeseed oil production line according to claim 1, characterized in that, Determining the static forming parameters of the oil cake in a single image of the oil cake includes: Superpixel segmentation is performed on the oil cake image to obtain multiple tile segmentation regions; The area features and brightness features of each tile segmentation region are extracted respectively, and the regional static forming parameters of each tile segmentation region are determined based on the area features and brightness features; The static forming parameters of the multiple tile segmentation regions are averaged to obtain the static forming parameters of the oil cake in the oil cake image.

3. The method for real-time monitoring of a rapeseed oil production line according to claim 2, characterized in that, Determining the static forming parameters of a single tile segmentation region includes: Based on the area characteristics of the tile segmentation region, a first local forming probability is determined; wherein, the first local forming probability is used to characterize the local forming probability of the oil cake in the tile segmentation region determined based on the area characteristics; Based on the brightness characteristics of the tile segmentation region, a second local forming probability is determined; wherein, the second local forming probability is used to characterize the local forming probability of the oil cake in the tile segmentation region determined based on the brightness characteristics; Based on the area feature distribution of all the tile segmentation regions in the oil cake image where the tile segmentation region is located, a first weight value is determined; wherein, the first weight value is used to characterize the overall forming probability of the oil cake in the oil cake image determined based on the area feature distribution; A second weight value is determined based on the brightness feature distribution of all tile segmentation regions in the oil cake image where the tile segmentation region is located; wherein, the second weight value is used to characterize the overall forming probability of the oil cake in the oil cake image determined based on the brightness feature distribution; Based on the first weight value and the second weight value, the first local forming probability and the second local forming probability are weighted and summed to obtain the regional static forming parameters of the tile segmentation region.

4. The method for real-time monitoring of a rapeseed oil production line according to claim 1, characterized in that, The step of determining the dynamic forming parameters of the oil cake within the preset time period based on the static forming parameters corresponding to multiple images of the oil cake within the preset time period includes: Based on the static forming parameters corresponding to multiple images of fried dough cakes within the preset time period, a temporal difference index is determined between two adjacent images of fried dough cakes; wherein, the temporal difference index is used to characterize the degree of abrupt change in the static forming parameters between two images of fried dough cakes. Cluster the time-series difference indicators to identify anomalous clusters; Based on the aforementioned abnormal clusters, the preset time period is divided into multiple consecutive segments; Based on the morphological characteristics of the oil cakes within multiple consecutive segments, the dynamic forming parameters of the oil cakes within the preset time period are determined.

5. The real-time monitoring method for a rapeseed oil production line according to claim 4, characterized in that, The step of clustering the time-series difference indicators to identify anomalous clusters includes: Clustering is performed on multiple time-series difference indicators to obtain a first cluster and a second cluster; Abnormal clusters and normal clusters are determined in the first cluster and the second cluster; wherein, the cluster containing a smaller number of samples is the abnormal cluster, and the cluster containing a larger number of samples is the normal cluster.

6. The method for real-time monitoring of a rapeseed oil production line according to claim 4, characterized in that, The step of dividing the preset time period into multiple consecutive segments based on the abnormal clusters includes: Starting from the beginning of the preset time period, the clusters to which the time-series difference indicators belong are determined sequentially; When the temporal difference index belonging to the anomalous cluster is first determined, the previous moment of the temporal difference index is taken as the end point of the current continuous segment. The endpoint of the continuous segment is determined by taking the next moment of the time series difference index as the starting point of the next continuous segment; Repeat the above process until the entire preset time period has been traversed to obtain multiple consecutive segments.

7. The method for real-time monitoring of a rapeseed oil production line according to claim 4, characterized in that, The determination of dynamic forming parameters of the oil cake within the preset time period based on the morphological characteristics of the oil cake within multiple consecutive segments includes: For each of the continuous segments, the oil cake image at each time point within the continuous segment is segmented into tiles to obtain multiple tile regions; For each of the tile regions, the tile similarity between the tile region and each of the tile regions at the target time is calculated to obtain the morphological parameters of each tile region; wherein, the target time is used to characterize all subsequent times within the continuous segment corresponding to the time of the oil pancake image where the tile region is located; the morphological parameter is the mean tile similarity of the tile region; Based on similarity weighting, the morphological parameters of each tile region within the continuous segment are weighted and summed to obtain the dynamic forming parameters of the continuous segment; wherein, the similarity weighting is used to characterize the degree of continuous influence of the oil cake image where the tile region is located in the continuous segment; Based on the consistency weight of the oil cake shape, the dynamic forming parameters of each continuous segment are weighted, summed, and averaged to determine the dynamic forming parameters of the oil cake within the preset time period; wherein, the consistency weight is used to characterize the degree of consistency of the oil cake shape between two adjacent continuous segments.

8. The method for real-time monitoring of a rapeseed oil production line according to any one of claims 1-7, characterized in that, Determining the forming state of the oil cake based on the dynamic forming parameters includes: The dynamic forming parameters of the oil cake within the preset time period are normalized. If the normalized dynamic forming parameters are greater than the preset forming threshold, the forming state of the oil cake is determined to be formed. If the normalized dynamic forming parameters are not greater than the preset forming threshold, then the forming state of the oil cake is determined to be unformed.

9. The method for real-time monitoring of a rapeseed oil production line according to any one of claims 1-7, characterized in that, After acquiring the time-series image sequence of rapeseed oil cake during the operation of the rapeseed oil production line, the method further includes: The images of fried dough cakes in the time-series image sequence are preprocessed; wherein the preprocessing includes at least one of filtering, grayscale conversion and threshold segmentation.

10. A real-time monitoring system for a rapeseed oil production line, characterized in that, include: Image acquisition device and host computer, wherein: The image acquisition device is installed at the cake discharge port of the rapeseed oil production line to acquire real-time images of the oil cake and send the images of the oil cake to the host computer. The host computer is used to acquire a time-series image sequence of rapeseed cake during the operation of the rapeseed oil production line based on the oil cake images acquired by the image acquisition device; determine the static forming parameters of the oil cake in multiple oil cake images within a preset time period; wherein the static forming parameters are used to characterize the probability of oil cake formation in a single oil cake image at the time of acquisition; determine the dynamic forming parameters of the oil cake within the preset time period based on the static forming parameters corresponding to multiple oil cake images within the preset time period; wherein the dynamic forming parameters are used to characterize the cumulative reliability of the oil cake maintaining a formed state within the preset time period; determine the forming state of the oil cake based on the dynamic forming parameters; and trigger a maintenance alarm for the rapeseed oil production line when the forming state is unformed.

Images