Online monitoring system and method for edible oil quality based on multi-modal sensor

Abstract

The present application relates to the field of edible oil quality monitoring, and particularly relates to an online edible oil quality monitoring system and method based on a multi-modal sensor, which extracts the pressing state characteristics of the outflowing oil liquid corresponding to the pressing process; calculates the pressing quality representation value of the outflowing oil liquid to mark the outflowing oil liquid; further evaluates and analyzes the quality of the outflowing oil liquid; extracts the flow disorder characteristics of the outflowing oil liquid in the entire pressing process to determine whether it meets the flow deviation fault tolerance requirement; according to the splashing height of the outflowing oil liquid entering the oil receiving container, in combination with the retention time and accumulation area of the foam in the oil receiving container, the quality deviation representation coefficient of the corresponding crude oil product is evaluated, the quality deviation level is determined, the corresponding color is marked, and a warning prompt information is sent. The present application realizes multi-dimensional and comprehensive edible oil quality monitoring by real-time online monitoring, timely discovery of edible oil production abnormalities, and improvement of monitoring accuracy.

Description

Technical Field

[0001] This invention relates to the field of edible oil quality monitoring, and in particular to an online edible oil quality monitoring system and method based on multimodal sensors. Background Technology

[0002] Edible oil, as a daily necessity, has its quality and safety directly impacting public health. Real-time, accurate quality monitoring during edible oil production is crucial for ensuring food safety and production efficiency. Online edible oil quality monitoring technology based on multimodal sensors is a natural product responding to the rigid demands for food safety and adapting to the trend of industrial intelligentization. By integrating multiple sensor technologies and artificial intelligence algorithms, it aims to solve the lag problem of traditional offline detection methods, achieving real-time visualization, controllability, and intelligence of the production process, ultimately ensuring oil quality, improving production efficiency, and reducing risks and costs.

[0003] Chinese Patent Application Publication No. CN118817924A discloses a method and system for dynamic detection of oil samples based on multimodal information in the field of oil sample detection technology. The method includes acquiring multimodal signals from the oil sample, extracting chromatographic characteristic peaks and spectral absorption peaks, and normalizing the electrical parameter signals to obtain chromatographic features, spectral features, and electrical parameter features. These features are then input into a multimodal oil sample feature fusion network to obtain fused features. The fused features are then input into a dynamic detection model for the oil sample, generating an oil sample status report and an oil sample fault diagnosis report based on the optimal state sequence of the oil sample. The oil sample status report includes the oil sample chromatographic fingerprint, oil sample spectral absorption curve, and oil sample comprehensive health index. The oil sample fault diagnosis report includes an analysis of the causes of oil sample deterioration, an assessment of the oil sample deterioration level, and recommendations for oil sample maintenance measures.

[0004] However, the following problems still exist in the existing technology.

[0005] Relying solely on a single technology (such as near-infrared spectroscopy) to reflect chemical composition for quality assessment ignores the physical state of the oil itself during the production process, which is rather one-sided and results in insufficient comprehensiveness and low accuracy in quality monitoring. Summary of the Invention

[0006] Therefore, the present invention provides an online monitoring system and method for edible oil quality based on multimodal sensors, which overcomes the problem that the existing technology ignores the physical state of the oil itself during the edible oil production process, resulting in a one-sided view and insufficient comprehensiveness and accuracy of quality monitoring.

[0007] To achieve the above objectives, the present invention provides an online monitoring method for edible oil quality based on a multimodal sensor, comprising:

[0008] Production data from a single edible oil production process is collected, and the pressing state characteristics of the outflowing oil during the pressing process are extracted. The pressing state characteristics include the maximum color difference of the outflowing oil from several oil outlet pores and the area of ​​impurities covering the edge of the oil outlet pores.

[0009] The pressing quality characterization value of the outflowing oil is calculated based on the pressing state characteristics, so as to mark the outflowing oil;

[0010] Based on the marking results of the outflowing oil, the quality of the outflowing oil is evaluated and analyzed, including determining whether the outflowing oil meets the flow stability benchmark based on the oil outflow fluctuation characteristics corresponding to the outflowing oil. The oil outflow fluctuation characteristics include the frequency of flow interruption and the average duration of flow interruption.

[0011] Extract the flow disorder characteristics of the outflowing oil during the entire pressing process, determine whether it meets the flow deviation tolerance requirements, evaluate the quality deviation characterization coefficient of the corresponding crude oil product based on the splash height of the outflowing oil entering the oil receiving container, combined with the residence time and foam accumulation area of ​​the foam in the oil receiving container, and determine the quality deviation level of the crude oil product based on the quality deviation characterization coefficient, mark it with the corresponding color and issue a warning message.

[0012] The flow disorder characteristics include the frequency of flow velocity reduction and the average duration of the reduction.

[0013] Furthermore, the process of calculating the pressing quality characterization value of the outflowing oil based on the pressing state characteristics includes:

[0014] The ratio of the maximum color difference of the oil flowing out of several oil outlet pores to the maximum color difference threshold is used as the first pressing quality characteristic.

[0015] The ratio of the edge impurity coverage area of ​​the oil outlet pores to the edge impurity coverage area threshold is used as the second pressing quality characteristic.

[0016] The sum of the first pressing quality characteristic and the second pressing quality characteristic is taken as the pressing quality characterization value.

[0017] Further, the outflowing oil is marked, including:

[0018] If the pressing quality characterization value of the outflowing oil is greater than or equal to the pressing quality characterization threshold, then the outflowing oil is marked.

[0019] Furthermore, based on the labeling results of the outflowing oil, the quality of the outflowing oil is evaluated and analyzed, including:

[0020] If any outflowing oil is marked, the quality of the outflowing oil is evaluated and analyzed.

[0021] Further, determining whether the outflowing oil meets the flow stability criteria includes:

[0022] If the frequency of outflow interruption of the outflowing oil is greater than the outflow interruption frequency threshold, and the average duration of the outflow interruption is greater than the average duration threshold, then the outflowing oil is determined to not meet the flow stability benchmark.

[0023] Further, determining whether the flow deviation tolerance requirements are met includes:

[0024] If the frequency of the outflowing oil velocity decreases is greater than the velocity decrease frequency threshold, or / and the average decrease duration is greater than the average decrease duration threshold, then it is determined that the flow deviation tolerance requirement is not met.

[0025] Furthermore, the process of evaluating the corresponding quality deviation characterization coefficient of the crude oil product includes:

[0026] The ratio of the splash height of the outflowing oil into the oil receiving container to the splash height threshold is used as the first quality deviation feature;

[0027] The sum of the ratio of foam residence time to residence time threshold and the ratio of foam accumulation area to foam accumulation area threshold is used as the second quality deviation feature.

[0028] The first quality deviation feature and the second quality deviation feature are weighted and summed to determine the quality deviation characterization coefficient.

[0029] Further, the process of determining the quality deviation grade for the crude oil product based on the quality deviation characterization coefficient includes:

[0030] The correspondence between the quality deviation level and the predetermined quality deviation characterization coefficient range is preset;

[0031] Determine the range of quality deviation characterization coefficients to which the quality deviation characterization coefficients for crude oil products belong;

[0032] The quality deviation level corresponds one-to-one with the range of the quality deviation characterization coefficient.

[0033] Furthermore, the process of assigning appropriate color-coding and issuing warning messages includes:

[0034] The colors marked correspond one-to-one with the quality deviation levels;

[0035] The warning information includes the color of the crude oil product being marked.

[0036] Furthermore, a system for online monitoring of edible oil quality based on multimodal sensors is also provided, characterized in that it includes:

[0037] The production acquisition module is used to collect production data during a single production process of edible oil and extract the pressing state characteristics of the oil flowing out during the pressing process. The pressing state characteristics include the maximum color difference of the oil flowing out of several oil outlet pores and the area of ​​impurities covering the edge of the oil outlet pores.

[0038] A pressing analysis module, which is connected to the production acquisition module, is used to calculate the pressing quality characterization value of the outflowing oil based on the pressing state characteristics, so as to mark the outflowing oil.

[0039] A quality assessment module, connected to the pressing analysis module, is used to assess and analyze the quality of the effluent oil based on the marking results for the effluent oil. This includes determining whether the effluent oil meets the flow stability benchmark based on the oil flow fluctuation characteristics corresponding to the effluent oil. The oil flow fluctuation characteristics include the frequency of flow interruption and the average duration of flow interruption.

[0040] The depth assessment module, connected to the quality assessment module, is used to extract the flow disorder characteristics of the outflowing oil during the entire pressing process, determine whether it meets the flow deviation tolerance requirements, evaluate the quality deviation characterization coefficient of the corresponding crude oil product based on the splash height of the outflowing oil entering the oil receiving container, combined with the retention time and foam accumulation area of ​​the foam in the oil receiving container, and determine the quality deviation level of the crude oil product based on the quality deviation characterization coefficient, mark it with the corresponding color and issue a warning message.

[0041] The flow disorder characteristics include the frequency of flow velocity reduction and the average duration of the reduction.

[0042] Compared with existing technologies, this invention includes a production acquisition module for collecting production data during a single edible oil production process and extracting the pressing state characteristics of the outflowing oil. A pressing analysis module calculates the pressing quality characterization value of the outflowing oil based on the pressing state characteristics, thus marking the outflowing oil. A quality assessment module evaluates and analyzes the quality of the outflowing oil based on the marking results. A depth assessment module extracts the flow disorder characteristics of the outflowing oil throughout the pressing process, determines whether it meets the flow deviation tolerance requirements, and assesses the corresponding quality deviation characterization coefficient of the crude oil product based on the splash height of the outflowing oil entering the receiving container, combined with the retention time and foam accumulation area within the receiving container. Based on the quality deviation characterization coefficient, it determines the quality deviation level of the crude oil product, assigns corresponding colors, and issues early warning information. This invention, through real-time online monitoring, promptly detects abnormalities in edible oil production, achieving multi-dimensional and comprehensive edible oil quality monitoring and improving monitoring accuracy.

[0043] In particular, this invention considers the visually apparent characteristics of edible oil during the pressing process. It quantifies the degree of color difference between the oil flowing from different oil outlets on the pressing equipment by using the maximum color difference of the oil flowing from several oil outlet pores. Generally, under the same pressing conditions, the color of the oil flowing from each pore in the same batch of raw materials should be relatively consistent. Therefore, the maximum color difference between different oil outlet pores can reflect the uniformity of the pressing process. Furthermore, the area covered by impurities at the edge of the oil outlet pores is directly related to the purity of the oil, reflecting the content of impurities that may be mixed into the oil. The area covered by impurities at the edge of the oil outlet pores quantifies the size of the area covered by impurities, thus reflecting the degree of impact on the purity and quality of the finished oil. Therefore, this invention reflects the quality of the oil and the stability of the pressing process through two core dimensions: oil uniformity and oil purity. This achieves quantitative monitoring of the pressing process quality, provides data support for subsequent labeling of the flowing oil, and enables rapid screening of "problem oil" through labeling. This invention enables timely detection of edible oil production anomalies through real-time online monitoring, achieving multi-dimensional and comprehensive edible oil quality monitoring and improving the accuracy of monitoring.

[0044] In particular, this invention further analyzes the flow stability of the labeled outflowing oil after pressing, which directly affects the uniformity of the pressing process. Interruptions in flow can cause sudden pressure changes and temperature fluctuations in local pressing areas, leading to accelerated oil oxidation, flavor deterioration, or an increased risk of impurity contamination. By quantifying the frequency and average duration of flow interruptions, the degree of flow anomalies is captured to comprehensively assess whether the oil flow deviates from the flow stability benchmark, thus avoiding batch-to-batch quality differences caused by uneven flow. This provides proactive assurance of outflowing oil quality. Through real-time online monitoring, this invention promptly detects anomalies in edible oil production, achieving multi-dimensional and comprehensive edible oil quality monitoring and improving monitoring accuracy.

[0045] In particular, this invention considers not only the state of the outflowing oil deviating from the flow stability benchmark during the flow process, but also the static state of the crude oil product that has entered the receiving container. Based on the relevant flow disorder characteristics of the outflowing oil throughout the pressing process, it reflects the dynamic stability of the pressing process. Under normal circumstances, a decrease in the outflowing oil velocity may stem from uneven raw material supply, press pressure fluctuations, or localized pore blockage. The frequency and average duration of the decrease can quantify the frequency and duration of flow disorder. By determining whether the flow deviation tolerance requirements are met, overreaction to small, acceptable fluctuations can be avoided, while excessive disorder exceeding the tolerance range can be identified, thus defining clear boundaries for process stability. Furthermore, by linking the physical state and intrinsic quality of the crude oil product, the intuitiveness and accuracy of quality assessment are improved. The splash height of the outflowing oil primarily reflects the flow stability and impurity content of the oil after pressing. Therefore, the splash height of the outflowing oil quantifies the smoothness of oil flow and the degree of impurity entrainment, indirectly reflecting the uniformity of the pressing process and the initial purity of the oil. The retention time of foam in the oil receiving container is mainly related to the colloidal impurities and water content in the oil. These impurities have surface activity, which reduces the surface tension of the oil, making the foam less prone to breakage and prolonging the retention time. Therefore, the foam retention time quantifies the activity and content of colloidal impurities in the oil, reflecting the refining difficulty and original purity of the oil. The foam accumulation area in the oil receiving container mainly reflects the total amount of substances in the oil that easily generate foam. Therefore, the foam accumulation area reflects the processing adaptability and quality grade of the oil. Furthermore, the quality deviation characterization coefficient of the corresponding crude oil product is comprehensively quantified to quantify the severity of quality deviations in the oil product. This invention, through real-time online monitoring, can promptly detect abnormalities in edible oil production, achieving multi-dimensional and comprehensive edible oil quality monitoring and improving the accuracy of monitoring. Attached Figure Description

[0046] Figure 1 This is a schematic diagram illustrating the steps of an online monitoring method for edible oil quality based on a multimodal sensor, as described in an embodiment of the invention.

[0047] Figure 2 This is a logic diagram for marking the outflowing oil in an embodiment of the invention.

[0048] Figure 3 A logic diagram for determining whether the outflowing oil meets the flow stability criteria in an embodiment of the invention;

[0049] Figure 4 This is a logic diagram for determining whether the flow deviation tolerance requirement is met in an embodiment of the invention. Detailed Implementation

[0050] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0051] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0052] It should be noted that in the description of this invention, the terms "upper," "lower," "inner," etc., indicating the direction or positional relationship are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0053] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0054] Please see Figure 1 The diagram illustrates the steps of an online edible oil quality monitoring method based on a multimodal sensor according to an embodiment of the present invention. The method includes:

[0055] Step S1: Collect production data during a single production process of edible oil, and extract the pressing state characteristics of the oil flowing out during the pressing process. The pressing state characteristics include the maximum color difference of the oil flowing out of several oil outlet pores and the edge impurity coverage area of ​​the oil outlet pores.

[0056] Step S2: Calculate the pressing quality characterization value of the outflowing oil based on the pressing state characteristics, so as to mark the outflowing oil;

[0057] Step S3: Based on the marking results of the outflowing oil, evaluate and analyze the quality of the outflowing oil, including determining whether the outflowing oil meets the flow stability benchmark based on the oil outflow fluctuation characteristics corresponding to the outflowing oil. The oil outflow fluctuation characteristics include the frequency of flow interruption and the average duration of flow interruption.

[0058] Step S4: Extract the flow disorder characteristics of the outflowing oil during the entire pressing process, determine whether it meets the flow deviation tolerance requirements, evaluate the quality deviation characterization coefficient of the corresponding crude oil product based on the splash height of the outflowing oil entering the oil receiving container, combined with the retention time and foam accumulation area of ​​the foam in the oil receiving container, determine the quality deviation level of the crude oil product based on the quality deviation characterization coefficient, mark it with the corresponding color and issue a warning message.

[0059] The flow disorder characteristics include the frequency of flow velocity reduction and the average duration of the reduction.

[0060] Specifically, the production data includes pressing state characteristics, oil output fluctuation characteristics, flow disorder characteristics, splash height, foam retention time in the oil receiving container, and foam accumulation area, etc.

[0061] Specifically, there are no specific limitations on the acquisition method for the characteristics of the pressing state. An industrial color camera (e.g., 5 megapixels or higher, supporting RGB three-channel imaging) can be installed directly in front of or diagonally above the oil outlet of the pressing equipment, in conjunction with a ring LED light source (to avoid shadow interference), to ensure clear imaging of the oil flow stream and the edge of each oil outlet. Based on the real-time captured images of the oil flow stream from each oil outlet, the oil region is extracted using an image segmentation algorithm, the average color value of each oil region is determined, and the maximum difference between the corresponding average color values ​​of each oil outlet is calculated, thereby obtaining the maximum color difference of the oil flowing out of several oil outlets.

[0062] The image of the edge region of the oil outlet pore is converted to grayscale. The boundary between impurities and the metal edge of the pore is identified by an edge detection algorithm (such as the Canny operator). The pixel ratio of impurities in the edge region is calculated. Combined with the actual physical size (the conversion relationship between pixels and millimeters is pre-calibrated by a calibration plate), the impurity coverage area of ​​the edge of the oil outlet pore is obtained.

[0063] Specifically, there is no specific limitation on the method of collecting and acquiring the oil flow characteristics. An infrared photoelectric sensor (with the transmitter and receiver located on opposite sides of the flow) can be installed on the oil flow path below the oil outlet orifice. When the oil flow is interrupted, the infrared light is unobstructed, and the receiver triggers a "flow interruption signal". Then, the number of times the "flow interruption signal" is triggered during a single pressing process can be counted to obtain the flow interruption frequency.

[0064] The process involves recording the duration of each flow interruption signal (the time from the start of the flow interruption to the resumption of oil flow) and calculating the average duration of all flow interruptions to obtain the average flow interruption duration. This will not be elaborated further.

[0065] Specifically, there is no specific limitation on the method of collecting and acquiring the flow disorder characteristics. A high-precision electromagnetic velocity meter can be installed in the pipeline from the pressing equipment to the oil receiving container. The normal flow rate range can be preset in advance. When the flow rate is lower than the lower limit value corresponding to the normal flow rate range, it is regarded as a flow rate reduction event. The number of flow rate reduction events during a single pressing process is counted, which is the flow rate reduction frequency.

[0066] Specifically, by recording the duration of each flow rate reduction event (the time from when the flow rate falls below the lower limit to when it recovers to the normal range), the average duration of all flow rate reduction events is calculated, which is the average reduction duration. The normal flow rate range can be based on historical production data corresponding to several edible oil production processes. The minimum value among several historical flow rates is taken as the lower limit of the normal flow rate range, and the maximum value is taken as the upper limit of the normal flow rate range. The normal flow rate range is a closed interval, which will not be elaborated further.

[0067] Specifically, there is no specific limitation on the method for acquiring the splash height. A laser rangefinder sensor can be installed directly above the oil receiving container, with its lens facing the oil inflow point. At the same time, a high-speed camera can be installed to record the instantaneous image of the splash. The distance between the laser rangefinder sensor and the oil surface can be monitored in real time. When the oil splashes, the distance value suddenly decreases (the highest point of the splash is closer to the sensor). The minimum distance value is recorded. Since the laser rangefinder sensor is fixedly installed, the splash height is obtained by subtracting the minimum distance value from the distance between the laser rangefinder sensor and the surface of the crude oil product in the oil receiving container. This will not be elaborated further.

[0068] The collection of foam retention time and foam accumulation area in the oil receiving container can be based on the surface image of the oil receiving container taken by the high-speed camera, combined with a machine learning model (such as a trained foam recognition CNN model to identify the foam area and calculate the pixel area of ​​the foam area to obtain the foam accumulation area).

[0069] The time difference between the time when the foam completely disappears and the time when the foam first appears is used as the foam retention time. If the foam does not completely disappear, the time difference between the time when the pressing process is completed and the time when the foam first appears is used as the foam retention time. This will not be elaborated further.

[0070] Specifically, the process of calculating the pressing quality characterization value of the outflowing oil based on the pressing state characteristics includes:

[0071] The ratio of the maximum color difference of the oil flowing out of several oil outlet pores to the maximum color difference threshold is used as the first pressing quality characteristic.

[0072] The ratio of the edge impurity coverage area of ​​the oil outlet pores to the edge impurity coverage area threshold is used as the second pressing quality characteristic.

[0073] The sum of the first pressing quality characteristic and the second pressing quality characteristic is taken as the pressing quality characterization value.

[0074] In this embodiment, the purpose of setting the maximum color difference threshold and the edge impurity coverage area threshold is to characterize situations where the oil quality is likely to have potential problems and the pressing process is unstable. Historical production data corresponding to several edible oil production processes completed using the same production equipment is obtained. Historical data on the maximum color difference of the outflowing oil from several oil outlet pores and historical data on the edge impurity coverage area of ​​the oil outlet pores are retrieved to calculate the average maximum color difference and the average edge impurity coverage area, which are then used as benchmark values ​​under normal conditions. Based on the purpose of setting the above two thresholds, the maximum color difference threshold is determined as the product of the average maximum color difference and the color deviation coefficient, and the edge impurity coverage area threshold is determined as the product of the average edge impurity coverage area and the coverage deviation coefficient. The color deviation coefficient is selected within the interval [1.1, 1.15], preferably 1.1 in practice, and the coverage deviation coefficient is selected within the interval [1.1, 1.2], preferably 1.1 in practice.

[0075] Specifically, this invention considers the visually apparent characteristics during the edible oil pressing process. It quantifies the degree of color difference between the oil flowing from different oil outlets on a pressing device by measuring the maximum color difference of the oil flowing from several oil outlet pores. Generally, under the same pressing conditions, the color of the oil flowing from each pore should be relatively consistent for the same batch of raw materials. Therefore, the maximum color difference between different oil outlet pores can reflect the uniformity of the pressing process (such as raw material distribution and pressure stability). If the color difference between the oil flowing from each oil outlet pore is too large, it may indicate localized issues. Abnormal pressing (such as overheating or under-pressing of raw materials in a certain area) affects the overall quality stability of the oil. Furthermore, the area of ​​impurities covering the edges of the oil outlet pores is directly related to the purity of the oil, reflecting the content of impurities that may have entered the oil. The area of ​​impurities covering the edges of the oil outlet pores quantifies the size of the area covered by impurities (such as raw material residue). A larger coverage area indicates that raw material residue is more easily discharged with the oil during the pressing process, leading to increased impurities in the oil, increasing the difficulty of subsequent filtration and refining, and ultimately potentially affecting the purity and quality of the finished oil. Therefore, this invention reflects the quality of the oil and the stability of the pressing process through two core dimensions: oil homogeneity and oil purity. This enables quantitative monitoring of the pressing process quality, providing data support for subsequent labeling of the discharged oil, and facilitating the rapid screening of "problem oil." This invention, through real-time online monitoring, promptly detects abnormalities in edible oil production, achieving multi-dimensional and comprehensive edible oil quality monitoring and improving the accuracy of monitoring.

[0076] Specifically, please refer to Figure 2 As shown, this is a logic decision diagram for marking the outflowing oil in an embodiment of the present invention. Marking the outflowing oil includes:

[0077] If the pressing quality characterization value of the outflowing oil is greater than or equal to the pressing quality characterization threshold, then the outflowing oil is marked.

[0078] If the pressing quality characterization value of the outflowing oil is less than the pressing quality characterization threshold, then there is no need to mark the outflowing oil.

[0079] The pressing quality characterization threshold is predetermined. The pressing quality characterization value calculated when the maximum difference in color of the oil flowing out of several oil outlet pores is equal to the maximum difference in color threshold, and the edge impurity coverage area of ​​the oil outlet pores is equal to the edge impurity coverage area threshold, is determined as the pressing quality characterization threshold.

[0080] Specifically, based on the labeling results of the outflowing oil, the quality of the outflowing oil is evaluated and analyzed, including:

[0081] If any outflowing oil is marked, the quality of the outflowing oil is evaluated and analyzed.

[0082] Specifically, please refer to Figure 3 As shown, this is a logic diagram for determining whether the outflowing oil meets the flow stability criterion in an embodiment of the present invention. Determining whether the outflowing oil meets the flow stability criterion includes:

[0083] If the frequency of outflow interruption of the outflowing oil is greater than the outflow interruption frequency threshold, and the average duration of the outflow interruption is greater than the average duration threshold, then the outflowing oil is determined to not meet the flow stability benchmark.

[0084] In this embodiment, the purpose of setting the flow interruption frequency threshold and the average flow interruption duration threshold is to characterize the situation where the deviation of the flow stability of the outflowing oil is relatively serious. By acquiring historical production data corresponding to several edible oil production processes completed by the same production equipment, the historical data of the outflowing oil flow interruption frequency and the corresponding historical data of the average flow interruption duration are called to solve for the average flow interruption frequency and the average flow interruption duration, and these are used as the benchmark values ​​under normal conditions. Based on the purpose of setting the above two thresholds, the flow interruption frequency threshold is determined as the product of the average flow interruption frequency and the first deviation coefficient, and the average flow interruption duration threshold is determined as the product of the average flow interruption duration and the second deviation coefficient. The first deviation coefficient is selected in the interval [1.2, 1.3], preferably 1.2 in the implementation, and the second deviation coefficient is selected in the interval [1.1, 1.15], preferably 1.1 in the implementation.

[0085] Specifically, this invention further analyzes the flow stability of the labeled outflowing oil after pressing, which directly affects the uniformity of the pressing process. Flow interruption (a brief cessation of oil flow) can cause sudden pressure changes and temperature fluctuations in local pressing areas (e.g., prolonged contact between the raw material and the press during an interruption may lead to localized overheating), thereby accelerating oil oxidation, deteriorating flavor, or increasing the risk of impurity contamination. By quantifying the frequency and average duration of flow interruptions, the degree of flow anomalies is captured to comprehensively assess whether the oil flow deviates from the flow stability benchmark, avoiding batch-to-batch oil quality differences caused by uneven flow (e.g., some areas of oil being overly scorched, others having excessive impurities). This achieves proactive assurance of outflowing oil quality. Through real-time online monitoring, this invention promptly detects anomalies in edible oil production, achieving multi-dimensional and comprehensive edible oil quality monitoring and improving monitoring accuracy.

[0086] Specifically, please refer to Figure 4 As shown, this is a logic diagram for determining whether the flow deviation tolerance requirement is met according to an embodiment of the present invention. The determination of whether the flow deviation tolerance requirement is met includes:

[0087] If the frequency of the outflowing oil velocity decreases is greater than the velocity decrease frequency threshold, or / and the average decrease duration is greater than the average decrease duration threshold, then it is determined that the flow deviation tolerance requirement is not met.

[0088] In this embodiment, the purpose of setting the flow rate reduction frequency threshold and the average reduction duration threshold is to characterize the situation where the outflowing liquid has a relatively serious flow disorder. By acquiring historical production data corresponding to several edible oil production processes completed by the same production equipment, the historical data of the outflowing oil flow rate reduction frequency and the corresponding historical data of the average reduction duration are called to solve for the average flow rate reduction frequency and the average reduction duration, and these are used as the benchmark values ​​under normal conditions. Based on the purpose of setting the above two thresholds, the flow rate reduction frequency threshold is determined as the product of the average flow rate reduction frequency and the first offset coefficient, and the average reduction duration threshold is determined as the product of the average reduction duration and the second offset coefficient. The first offset coefficient is selected in the interval [1.2, 1.3], preferably 1.2 in the implementation, and the second offset coefficient is selected in the interval [1.15, 1.2], preferably 1.15 in the implementation.

[0089] Specifically, the process of evaluating the quality deviation characterization coefficient of the corresponding crude oil product includes:

[0090] The ratio of the splash height of the outflowing oil into the oil receiving container to the splash height threshold is used as the first quality deviation feature;

[0091] The sum of the ratio of foam residence time to residence time threshold and the ratio of foam accumulation area to foam accumulation area threshold is used as the second quality deviation feature.

[0092] The first quality deviation feature and the second quality deviation feature are weighted and summed to determine the quality deviation characterization coefficient.

[0093] Specifically, the residence time and accumulation area of ​​foam in the oil receiving container directly correspond to the content of substances such as phospholipids, free fatty acids, colloidal substances, and water in the oil. These substances are also hard defects in crude oil quality, affecting not only the stability of the oil (easy to oxidize and rancid) but also being the main targets of subsequent refining processes (degumming, deacidification, and deimpurification). Both reflect the types and contents of inherent impurities in the crude oil product, which are the core factors determining the quality of crude oil. While a high splash height may increase the risk of oxidation (large contact area with air) and the contamination of impurities (such as contaminants on the container wall), these effects are more process-oriented. For example, the risk of oxidation can be mitigated by shortening the subsequent storage time, and the contaminated external impurities can be removed by simple processes such as filtration. The impact on the intrinsic quality of the crude oil product (such as the content of colloidal impurities) is indirect, mainly reflecting the stability of the oil flow and the resulting external effects. Therefore, in implementation, the second quality deviation characteristic calculated based on the residence time of foam in the oil receiving container and the foam accumulation area is given a higher weighting coefficient, set to 0.6, and the weighting coefficient of the first quality deviation characteristic calculated based on the splash height of the outflowing oil entering the oil receiving container is set to 0.4.

[0094] In this embodiment, the purpose of setting the splash height threshold, residence time threshold, and foam accumulation area threshold is to characterize situations where the oil flow is less stable and the oil is less stable. By acquiring historical production data corresponding to several edible oil production processes completed using the same production equipment, and by calling historical data on the splash height, residence time, and foam accumulation area of ​​the outflowing oil entering the receiving container, the average splash height, average residence time, and average foam accumulation area are calculated and used as the baseline values ​​under normal conditions. Based on the purpose of setting the above three thresholds, the... The splash height threshold is determined as the product of the average splash height and the height deviation coefficient; the dwell time threshold is determined as the product of the average dwell time and the dwell time deviation coefficient; and the foam accumulation area threshold is determined as the product of the average foam accumulation area and the accumulation deviation coefficient. The height deviation coefficient is selected within the interval [1.02, 1.05], preferably 1.02 in practice; the dwell time deviation coefficient is selected within the interval [1.15, 1.2], preferably 1.15 in practice; and the accumulation deviation coefficient is selected within the interval [1.2, 1.25], preferably 1.2 in practice.

[0095] Specifically, this invention considers both the state of the outflowing oil deviating from the flow stability benchmark during the flow process and the static state of the crude oil product already in the receiving container. Based on the relevant flow disorder characteristics of the outflowing oil throughout the pressing process—namely, the frequency and average duration of flow rate reduction—it reflects the dynamic stability of the pressing process. Under normal circumstances, a decrease in the outflowing oil's flow rate may stem from uneven raw material supply, press pressure fluctuations, or localized pore blockage. The frequency and average duration of the reduction quantify the frequency and duration of flow disorder. By determining whether the flow deviation tolerance requirements are met, it avoids overreacting to minor, acceptable fluctuations while identifying excessive disorder exceeding the tolerance range (such as frequent and prolonged sudden drops in flow rate), thus defining clear boundaries for process stability. Furthermore, it correlates the physical state and intrinsic quality of the crude oil product, improving the intuitiveness and accuracy of quality assessment. The splash height of the outflowing oil primarily reflects the flow stability and impurity content of the oil after pressing. For example, a higher splash height indicates greater pressure fluctuations and unstable flow velocity during the oil's outflow process, potentially accompanied by more solid impurities. These impurities alter the oil's fluidity, leading to increased splashing upon impact with the receiving container. Therefore, quantifying the smoothness of oil flow and the degree of impurity entrainment through the splash height indirectly reflects the uniformity of the pressing process and the initial purity of the oil. The residence time of foam in the receiving container is mainly related to the colloidal impurities (such as phospholipids, proteins, and mucilage) and moisture content in the oil. These impurities have surface activity, reducing the surface tension of the oil, making the foam less prone to breakage and prolonging its residence time. Therefore, quantifying the activity and content of colloidal impurities in the oil through foam residence time reflects the refining difficulty (more impurities mean more complex subsequent degumming and impurity removal processes) and original purity (high-purity oil produces less foam that disappears quickly). The area of ​​foam accumulation in the receiving container mainly reflects the total amount of substances in the oil that easily generate foam (such as phospholipids, free fatty acids, and moisture). The larger the foam accumulation area, the higher the content of this substance, the worse the oil stability (easier to oxidize and rancid), and the greater the load on subsequent refining processes. Therefore, the foam accumulation area reflects the processing adaptability and quality grade of the oil. Furthermore, the quality deviation characterization coefficient of the corresponding crude oil product is used to comprehensively quantify the severity of quality deviations in the oil product. This invention, through real-time online monitoring, can promptly detect anomalies in edible oil production, achieving multi-dimensional and comprehensive edible oil quality monitoring and improving monitoring accuracy.

[0096] Specifically, the process of determining the quality deviation grade for the crude oil product based on the quality deviation characterization coefficient includes:

[0097] The correspondence between the quality deviation level and the predetermined quality deviation characterization coefficient range is preset;

[0098] Determine the range of quality deviation characterization coefficients to which the quality deviation characterization coefficients for crude oil products belong;

[0099] The quality deviation level corresponds one-to-one with the range of the quality deviation characterization coefficient.

[0100] In this embodiment, the quality deviation level of the crude oil product is determined in the following manner:

[0101] The quality deviation characterization coefficient is divided into three preset intervals, and three levels of quality deviation are set.

[0102] If the quality deviation characterization coefficient of the crude oil product is within the first preset range (0, 1.67), then it corresponds to the first quality deviation level.

[0103] If the quality deviation characterization coefficient of the crude oil product is within the second preset range (1.67, 1.74), then it corresponds to the second quality deviation level.

[0104] If the quality deviation characterization coefficient of the crude oil product is within the third preset range (1.74, +∞), then it corresponds to the third quality deviation level.

[0105] Specifically, the process of assigning appropriate color-coding and issuing warning messages includes:

[0106] The colors marked correspond one-to-one with the quality deviation levels;

[0107] The warning information includes the color of the crude oil product being marked.

[0108] In this embodiment, the correspondence between quality deviation levels and labeled colors is determined as follows: the first quality deviation level corresponds to green; the second quality deviation level corresponds to yellow; and the third quality deviation level corresponds to red.

[0109] Specifically, a system for online monitoring of edible oil quality based on a multimodal sensor is also provided, characterized in that it includes:

[0110] The production acquisition module is used to collect production data during a single production process of edible oil and extract the pressing state characteristics of the oil flowing out during the pressing process. The pressing state characteristics include the maximum color difference of the oil flowing out of several oil outlet pores and the area of ​​impurities covering the edge of the oil outlet pores.

[0111] A pressing analysis module, which is connected to the production acquisition module, is used to calculate the pressing quality characterization value of the outflowing oil based on the pressing state characteristics, so as to mark the outflowing oil.

[0112] A quality assessment module, connected to the pressing analysis module, is used to assess and analyze the quality of the effluent oil based on the marking results for the effluent oil. This includes determining whether the effluent oil meets the flow stability benchmark based on the oil flow fluctuation characteristics corresponding to the effluent oil. The oil flow fluctuation characteristics include the frequency of flow interruption and the average duration of flow interruption.

[0113] The depth assessment module, connected to the quality assessment module, is used to extract the flow disorder characteristics of the outflowing oil during the entire pressing process, determine whether it meets the flow deviation tolerance requirements, evaluate the quality deviation characterization coefficient of the corresponding crude oil product based on the splash height of the outflowing oil entering the oil receiving container, combined with the retention time and foam accumulation area of ​​the foam in the oil receiving container, and determine the quality deviation level of the crude oil product based on the quality deviation characterization coefficient, mark it with the corresponding color and issue a warning message.

[0114] The flow disorder characteristics include the frequency of flow velocity reduction and the average duration of the reduction.

[0115] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for online monitoring of edible oil quality based on multimodal sensors, characterized in that, include: Production data from a single edible oil production process is collected, and the pressing state characteristics of the outflowing oil during the pressing process are extracted. The pressing state characteristics include the maximum color difference of the outflowing oil from several oil outlet pores and the area of ​​impurities covering the edge of the oil outlet pores. The pressing quality characterization value of the outflowing oil is calculated based on the pressing state characteristics, so as to mark the outflowing oil; Based on the labeling results of the outflowing oil, the quality of the outflowing oil is evaluated and analyzed. This includes determining whether the outflowing oil meets the flow stability benchmark based on the oil outflow fluctuation characteristics corresponding to the outflowing oil, wherein the oil outflow fluctuation characteristics include the frequency of flow interruption and the average duration of flow interruption. If the outflowing oil does not meet the flow stability benchmark, the flow disorder characteristics of the outflowing oil during the entire pressing process are extracted to determine whether it meets the flow deviation tolerance requirements. If the flow deviation tolerance requirement is not met, the quality deviation characterization coefficient of the corresponding crude oil product is evaluated based on the splash height of the outflowing oil entering the oil receiving container, combined with the residence time of the foam in the oil receiving container and the foam accumulation area. Based on the quality deviation characterization coefficient, the quality deviation level of the crude oil product is determined, and corresponding color marking is performed and a warning message is issued. The flow disorder characteristics include the frequency of flow velocity reduction and the average duration of reduction; The process of calculating the pressing quality characterization value of the outflowing oil based on the pressing state characteristics includes: The ratio of the maximum color difference of the oil flowing out of several oil outlet pores to the maximum color difference threshold is used as the first pressing quality characteristic. The ratio of the edge impurity coverage area of ​​the oil outlet pores to the edge impurity coverage area threshold is used as the second pressing quality characteristic. The sum of the first pressing quality characteristic and the second pressing quality characteristic is taken as the pressing quality characterization value; The process of evaluating the quality deviation characterization coefficient of the corresponding crude oil product includes: The ratio of the splash height of the outflowing oil into the oil receiving container to the splash height threshold is used as the first quality deviation feature; The sum of the ratio of foam residence time to residence time threshold and the ratio of foam accumulation area to foam accumulation area threshold is used as the second quality deviation feature. The first quality deviation feature and the second quality deviation feature are weighted and summed to determine the quality deviation characterization coefficient.

2. The online monitoring method for edible oil quality based on multimodal sensors according to claim 1, characterized in that, Marking the outflowing oil includes: If the pressing quality characterization value of the outflowing oil is greater than or equal to the pressing quality characterization threshold, then the outflowing oil is marked.

3. The online monitoring method for edible oil quality based on multimodal sensors according to claim 2, characterized in that, Based on the labeling results of the outflowing oil, the quality of the outflowing oil is evaluated and analyzed, including: If any outflowing oil is marked, the quality of the outflowing oil is evaluated and analyzed.

4. The method for online monitoring of edible oil quality based on multimodal sensors according to claim 1, characterized in that, Determining whether the outflowing oil meets the flow stability criteria includes: If the frequency of outflow interruption of the outflowing oil is greater than the outflow interruption frequency threshold, and the average duration of the outflow interruption is greater than the average duration threshold, then the outflowing oil is determined to not meet the flow stability benchmark.

5. The online monitoring method for edible oil quality based on multimodal sensors according to claim 1, characterized in that, Determining whether the flow deviation tolerance requirement is met includes: If the frequency of the outflowing oil velocity decreases is greater than the velocity decrease frequency threshold, or / and the average decrease duration is greater than the average decrease duration threshold, then it is determined that the flow deviation tolerance requirement is not met.

6. The online monitoring method for edible oil quality based on multimodal sensors according to claim 1, characterized in that, The process of determining the quality deviation grade for the crude oil product based on the quality deviation characterization coefficient includes: The correspondence between the quality deviation level and the predetermined quality deviation characterization coefficient range is preset; Determine the range of quality deviation characterization coefficients to which the quality deviation characterization coefficients for crude oil products belong; The quality deviation level corresponds one-to-one with the range of the quality deviation characterization coefficient.

7. The online monitoring method for edible oil quality based on multimodal sensors according to claim 1, characterized in that, The process of assigning appropriate color-coding and issuing warning messages includes: The colors marked correspond one-to-one with the quality deviation levels; The warning information includes the color of the crude oil product being marked.

8. A system for online monitoring of edible oil quality based on a multimodal sensor as described in any one of claims 1-7, characterized in that, include, The production acquisition module is used to collect production data during a single production process of edible oil and extract the pressing state characteristics of the oil flowing out during the pressing process. The pressing state characteristics include the maximum color difference of the oil flowing out of several oil outlet pores and the area of ​​impurities covering the edge of the oil outlet pores. A pressing analysis module, which is connected to the production acquisition module, is used to calculate the pressing quality characterization value of the outflowing oil based on the pressing state characteristics, so as to mark the outflowing oil. A quality assessment module, connected to the pressing analysis module, is used to assess and analyze the quality of the effluent oil based on the labeling results for the effluent oil. This includes determining whether the outflowing oil meets the flow stability benchmark based on the oil outflow fluctuation characteristics corresponding to the outflowing oil, wherein the oil outflow fluctuation characteristics include the frequency of flow interruption and the average duration of flow interruption. The depth assessment module, connected to the quality assessment module, is used to extract the flow disorder characteristics of the outflowing oil during the entire pressing process, determine whether it meets the flow deviation tolerance requirements, evaluate the quality deviation characterization coefficient of the corresponding crude oil product based on the splash height of the outflowing oil entering the oil receiving container, combined with the retention time and foam accumulation area of ​​the foam in the oil receiving container, and determine the quality deviation level of the crude oil product based on the quality deviation characterization coefficient, mark it with the corresponding color and issue a warning message. The flow disorder characteristics include the frequency of flow velocity reduction and the average duration of the reduction.

Images