A method and system for abnormal alarm of feed additive production and processing equipment

Abstract

The present invention discloses an abnormal alarm method and system for feed additive production and processing equipment, which relates to the technical field of abnormal alarm for production equipment. The invention can comprehensively monitor multiple key parameters of equipment operation, including but not limited to temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S. This comprehensive monitoring system can capture subtle changes in the equipment's operating status, especially those early signals that indicate potential failures, thereby achieving intelligent early warning, and constructing an adhesion influencing factor A for adhesion problems in key parts of the equipment. yx And evaluate, trigger the first adhesion warning information and the second adhesion warning information. The implementation of this multi-level warning mechanism allows enterprises to adjust maintenance strategies according to the warning level, avoiding unnecessary excessive maintenance, while also ensuring that timely measures are taken before the adhesion problem worsens, thereby finding the best balance between maintenance costs and equipment performance.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal alarm of production equipment, and in particular to an abnormal alarm method and system for production equipment of feed additive production and processing. Background Art

[0002] With the innovation of automation technology, feed additive production and processing equipment is a vital component in the feed production industry, used to manufacture various types of feed additives to improve the nutritional value of feed, promote animal growth, increase feed conversion rate and prevent diseases. These equipment usually need to have high precision, high efficiency and automation level to ensure the quality of additives and the safety of the production process.

[0003] There are some significant omissions and limitations in the abnormal alarm system of traditional feed additive production and processing equipment, which not only affect production efficiency, but also may pose a threat to product quality and production safety. Traditional abnormal alarm systems are often limited to monitoring a single or a few key parameters, such as temperature or pressure, while ignoring other important indicators in equipment operation, such as noise value, liquid level height, movement offset and cumulative running time. This limitation makes it impossible for the system to fully capture subtle changes in the operating status of the equipment, especially those early signals that indicate potential failures, resulting in untimely early warning of equipment failures, affecting production continuity and product quality. In addition, the status of key parts inside the equipment, such as stirring rods and cutting knives, is prone to accumulation of adhesions during the production of feed additives, which affects the operating efficiency and product quality of the equipment. Traditional adhesions are manually inspected, and the judgment criteria and cleaning degree of adhesions often rely on personal experience. Differences between different operators may lead to uneven maintenance effects. Some places are over-cleaned, resulting in unnecessary waste of resources, while others are not cleaned enough, leaving hidden dangers. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for abnormal alarm of feed additive production and processing equipment, which has solved the problems mentioned in the background technology.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for abnormal alarm of feed additive production and processing equipment, comprising the following steps:

[0008] Step 1: Divide the feed additive production area into several production areas according to the production process; the production area includes raw material processing area, mixing area, granulation area, drying area and packaging area;

[0009] Step 2: Set up several monitoring points in each area; install image acquisition equipment and sensor equipment at each monitoring point, collect and obtain the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S of the production equipment at several monitoring points in real time, and establish an abnormal monitoring data set;

[0010] Step 3: After filtering, denoising and format conversion of the abnormal monitoring data set, the parameter range of the production equipment under normal operating conditions is analyzed based on the historical data set, and the alarm threshold of each parameter is set;

[0011] The operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operation time S of the production equipment are compared with the preset alarm thresholds. When the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operation time S of the production equipment exceed the corresponding alarm threshold range, the corresponding warning information is triggered;

[0012] Step 4: Use the image acquisition device installed at each monitoring point to take real-time images of the stirring rod, cutting tool or roller surface in the production equipment to obtain a second image data set; after processing the second image data set, establish a defect recognition model, use an image processing algorithm to perform edge monitoring and morphological processing on the second image data set, and then extract the shape, area and position of the adhesion, evaluate the impact of the adhesion on the equipment operation efficiency, and construct the adhesion impact factor A for each production equipment. yx ;

[0013] Step 5: Adhesion influence factor A for each production equipment yx An assessment is performed to obtain a second assessment result, and corresponding warning information is triggered according to the second assessment result.

[0014] Preferably, several monitoring points are set up in each production area and arranged according to the key locations of the production equipment;

[0015] Each monitoring point should cover key production equipment and components, including stirring rods, mixing barrel walls, conveying pipelines and granulator molds; the following equipment should be installed at each monitoring point:

[0016] Temperature sensor, used to monitor the operating temperature T of production equipment;

[0017] Noise sensor, used to monitor the noise value Z during the operation of production equipment;

[0018] Liquid level sensor, used to monitor the liquid level height G of liquid raw materials or products;

[0019] Position sensor, used to monitor the position movement offset value P of the production equipment;

[0020] The operating time sensor is used to monitor the accumulated operating time S of the production equipment.

[0021] Preferably, the step three comprises:

[0022] S31, collecting historical data, analyzing the historical data, determining the parameter range of the equipment under normal operating conditions, and obtaining analysis results, including: normal operating temperature range, normal noise level, normal liquid level height, normal position offset range, and normal operating time range;

[0023] S32. According to the analysis results, set the initial alarm threshold of each parameter, wherein the initial alarm threshold is within the upper limit of the normal operation of the equipment; set the temperature threshold Noise Threshold Liquid level threshold Motion offset threshold and runtime thresholds

[0024]

[0025] S33, the operating temperature T of the production equipment is compared with the temperature threshold For comparison, when the operating temperature of the production equipment T> the temperature threshold Calculate the first temperature difference Generate a first temperature warning message, indicating that the temperature exceeds the preset temperature threshold range and there is a risk of damage to the production equipment, including overheating of the equipment due to excessive temperature, aging or burning of components;

[0026] The noise value Z and the noise threshold For comparison, when the noise value Z> noise threshold Calculate the first noise difference Generate the first noise warning information, indicating that the equipment has mechanical failure or component wear, and excessive noise indicates bearing damage, gear wear or insufficient lubrication;

[0027] The liquid level height G and the liquid level threshold For comparison, when the liquid level G> liquid level threshold Calculate the first level difference Generate the first liquid level warning information. Too high a liquid level may cause overflow or equipment damage. When the liquid level height G is less than the liquid level threshold Second level difference Generate a second liquid level warning message. If the liquid level is too low, the pump will run dry or the material supply will be insufficient.

[0028] The moving offset value P and the moving offset threshold For comparison, when the moving offset value P> moving offset threshold Calculate the first moving offset difference Generate the first collision warning information, the equipment position deviation leads to alignment error, product quality problems or equipment collision damage;

[0029] The cumulative running time S is compared with the running time threshold For comparison, when the cumulative running time S> running time threshold Calculate the first run time difference The first timeout warning information is generated, indicating that long-term operation will cause increased equipment wear, depletion of lubricating oil, or fatigue failure of components.

[0030] Preferably, the step 4 comprises:

[0031] S41. Install high-definition cameras or industrial cameras at each monitoring point to ensure coverage of key parts of the stirring rod, cutting tool and drum;

[0032] S42, setting the image acquisition frequency, including 5 frames per second or 3 frames per second, to acquire images of the surface of the production equipment in real time and construct a second image data set;

[0033] S43, filtering and denoising the second image data set, adjusting image contrast and brightness, and performing enhancement processing;

[0034] S44, use the Canny edge detection algorithm to identify the edges in the image, use morphological operations, including dilation, erosion, opening and closing operations to process the edge image, further separate the adhesion and background, and remove noise and small artifacts;

[0035] S45, identifying the contour of the adhesion by a contour monitoring algorithm, extracting the shape of the adhesion from the image, calculating the area of ​​the adhesion according to the contour, and determining the position of the adhesion in the image, and representing the position by a bounding box or an outer rectangle;

[0036] S46. Collect and annotate sample images, mark the shape, area, density and position features of different types of adhesions, construct training sets and test sets, use machine learning or convolutional neural network CNN to train the annotated sample images, establish a defect recognition model that can automatically identify and classify different types of adhesions, and use the second image data set as input to the defect recognition model to calculate and analyze: the adhesion influence factor A of each production equipment yx .

[0037] Preferably, the S46 includes:

[0038] S461, first calculate the relative adhesion area A: Among them, A total is the total area of ​​the adhesion, calculated by the image processing algorithm, A surface is the total surface area of ​​the equipment;

[0039] S462. Calculate adhesion density N md : N is the number of adhesions, which is obtained by counting the image processing algorithm;

[0040] S463, extract and calculate the shape factor C, comprehensively consider the circularity and aspect ratio parameters of the adhesion, and calculate the shape factor C:

[0041]

[0042] In the formula, A i is the area of ​​the ith adhesion, P i is the perimeter of the ith adhesion, L i and W i are the length and width of the ith adhesion, respectively;

[0043] S464. Comprehensively consider the distance from the center of the adhesion to the key components of the equipment and obtain the position factor D:

[0044]

[0045] Where, d i is the distance from the ith adhesion to the key part of the equipment, d max It is the maximum distance from any point on the surface of the equipment to the key part.

[0046] Preferably, the S46 further includes:

[0047] Adhesion influence factor A for each production equipment yx Calculated by the following steps:

[0048]

[0049] Where w1, w2, w3 and w4 represent the relative adhesion area A, adhesion density N, respectively. md , the weight values ​​of shape factor C and position factor D, and 0≤w1≤1, 0≤w2≤1, 0≤w3≤1, 0≤w4≤1, and w1+w2+w3+w4=1, d1 is represented as a correction constant.

[0050] Preferably, the adhesion influence factor A for each production equipment yx Conduct an assessment and obtain the second assessment results, including:

[0051] Setting a first impact threshold X1 and a second impact threshold X2, wherein the first impact threshold X1>the second impact threshold X2;

[0052] When the adhesion factor A yx < the second impact threshold X2, indicating that the adhesion state of the production equipment is normal and no warning is triggered;

[0053] When the second impact threshold X2 ≤ adhesion impact factor A yx ≤ the first impact threshold X1, indicating that the adhesion state of the production equipment is abnormal, triggering the first adhesion warning information, which is: "The equipment has a slight adhesion problem, please check and clean it";

[0054] When the adhesion factor A yx >The first impact threshold X1 indicates that the adhesion state of the production equipment is abnormal, triggering the second adhesion warning information. The second adhesion warning information is: "The equipment has serious adhesion problems, please check and clean it."

[0055] Preferably, the first temperature difference T1, the first noise difference Z1, the first liquid level difference G1, the second liquid level difference G2, the first movement offset difference P1, and the first operating time difference S1 are dimensionlessly processed, prioritized from high to low according to the size of the numerical values, and alarm processing is performed according to the corresponding first temperature warning information, first noise warning information, first liquid level warning information, second liquid level warning information, first collision warning information, and first timeout warning information.

[0056] Preferably, a first strategy is generated according to the first temperature warning information: immediately and automatically reduce the operating power of the production equipment by 30% or start the cooling system, and at the same time send an alarm to the maintenance team for on-site inspection to ensure that the heat dissipation equipment operates normally, and record the temperature abnormality history;

[0057] Generate a second strategy based on the first noise warning information: stop the machine for inspection, check the bearings and gear components, and replace the damaged components and readjust the lubrication system when the bearings and gear components are damaged;

[0058] Generate a third strategy based on the first liquid level warning information: automatically adjust the feed rate or discharge excess liquid to prevent overflow;

[0059] Generate a fourth strategy based on the second liquid level warning information: start the standby pump or manually increase the supply to prevent production interruption due to insufficient supply;

[0060] Generate a fifth strategy based on the first collision warning information: emergency stop the relevant equipment, use the positioning system to correct the position of the equipment, check and adjust the guidance device or alignment mechanism, and restart after confirmation;

[0061] Generate the sixth strategy based on the first timeout warning information: planned shutdown maintenance, perform regular maintenance such as changing lubricating oil, checking fasteners and wearing parts;

[0062] Generate the seventh strategy based on the first adhesion warning: initiate cleaning procedures, including using an automatic cleaning system or arranging manual cleaning;

[0063] The eighth strategy is generated based on the second adhesion warning: shut down the entire machine for cleaning and analyze the cause of adhesion. If it is a problem with the raw material formula, process parameters or equipment design, adjustments must be made before restarting production.

[0064] An abnormal alarm system for feed additive production and processing equipment, comprising:

[0065] The area division unit is used to divide the feed additive production area into several production areas according to the production process; the production area includes the raw material processing area, mixing area, granulation area, drying area and packaging area;

[0066] The first monitoring unit is used to install image acquisition equipment and sensor equipment at each monitoring point, collect and obtain the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S of the production equipment at several monitoring points in real time, and establish an abnormal monitoring data set;

[0067] The first processing unit performs data filtering, denoising and format conversion on the abnormal monitoring data set, analyzes the parameter range of the production equipment under normal operating conditions based on the historical data set, and sets the alarm threshold of each parameter;

[0068] The first evaluation unit compares the operating temperature T, noise value Z, liquid level height G, mobile offset value P and cumulative operating time S of the production equipment with the preset alarm threshold. When the operating temperature T, noise value Z, liquid level height G, mobile offset value P and cumulative operating time S of the production equipment exceed the corresponding alarm threshold range, the corresponding warning information is triggered.

[0069] The image acquisition unit uses an image acquisition device installed at each monitoring point to capture the device image of the stirring rod, cutting tool or drum surface in the production equipment in real time to obtain a second image data set;

[0070] Establish a defect recognition model analysis unit, establish a defect recognition model, use image processing algorithms to perform edge monitoring and morphological processing on the second image data set, extract the shape, area and position of the adhesion, evaluate the impact of adhesion on equipment operation efficiency, and construct the adhesion impact factor A for each production equipment yx ;

[0071] The second evaluation unit is used to evaluate the adhesion influence factor A of each production equipment. yx Performing an assessment to obtain a second assessment result, and triggering corresponding warning information according to the second assessment result;

[0072] The priority unit performs dimensionless processing on the first temperature difference T1, the first noise difference Z1, the first liquid level difference G1, the second liquid level difference G2, the first movement offset difference P1, and the first operation time difference S1, and prioritizes them from high to low according to the values, and performs corresponding priority sorting and alarm processing on the corresponding warning information;

[0073] The strategy generating unit generates a corresponding strategy according to the first temperature warning information, the first noise warning information, the first liquid level warning information, the second liquid level warning information, the first collision warning information, the first timeout warning information, the first adhesion warning information and the second adhesion warning information.

[0074] (III) Beneficial effects

[0075] The present invention provides a method and system for abnormal alarm of feed additive production and processing equipment. It has the following beneficial effects:

[0076] (1) By subdividing the production area and setting multiple monitoring points, combined with real-time data collection of image acquisition and sensor equipment, the present invention can comprehensively monitor multiple key parameters of equipment operation, including but not limited to temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S. This comprehensive monitoring system can capture subtle changes in the equipment's operating status, especially those early signals that indicate potential failures, thereby achieving intelligent early warning, preventing equipment failures in advance, and ensuring production continuity and product quality.

[0077] (2) This feed additive production and processing equipment abnormal alarm method and system, for the problem of adhesion in key parts of the equipment (such as stirring rods and cutting tools), can automatically identify and quantify the shape, area and position of adhesion through real-time monitoring of image acquisition equipment and image processing algorithm, and construct adhesion influence factor A. yx This innovation can not only evaluate the impact of adhesion on equipment operating efficiency, but also the adhesion impact factor A yx The calculation provides a scientific basis for maintenance decisions and avoids over-maintenance or under-maintenance. yx By evaluating the triggered first adhesion warning information and the second adhesion warning information, the implementation of this multi-level warning mechanism allows enterprises to adjust maintenance strategies according to the warning level, avoiding unnecessary excessive maintenance, while also ensuring that timely measures are taken before the adhesion problem worsens, thereby finding the best balance between maintenance costs and equipment performance.

[0078] (3) The processed first temperature difference T1, the first noise difference Z1, the first liquid level difference G1, the second liquid level difference G2, the first movement offset difference P1, and the first operation time difference S1 are sorted according to the numerical value, which can automatically identify the abnormal situation that has the most urgent impact on the operation of the equipment, ensuring that urgent problems can be handled as a priority, ensuring that within the limited maintenance capacity and time, the abnormalities that are most likely to cause production interruption or equipment damage are resolved first, thereby improving the efficiency and pertinence of maintenance response.

[0079] (4) The system achieves real-time and accurate control of the operating status of production equipment by finely dividing and comprehensively monitoring the production area. The system integrates the monitoring and analysis of multi-dimensional data such as temperature, noise, liquid level, movement deviation and operating time, and combines image acquisition and defect recognition technology. It can not only immediately detect and warn of equipment abnormalities such as overheating, excessive noise, abnormal liquid level, movement deviation and adhesion problems, but also prioritize them according to the severity of the warning information to ensure that key issues are handled first. The system improves the accuracy and effectiveness of warnings by dynamically setting alarm thresholds and using historical data to analyze the normal operating parameter range of equipment. In particular, the analysis of equipment images through image processing algorithms can quantitatively evaluate the impact of adhesions on equipment operating efficiency, construct adhesion influence factors, and provide an intuitive basis for equipment maintenance. In addition, the emergency strategies automatically generated by the system can guide the maintenance team to respond quickly and take corresponding measures, such as adjusting power, starting the cooling system, stopping for inspection, automatically adjusting feed or discharge, starting standby pumps, emergency stopping to correct equipment position, performing regular maintenance or comprehensive cleaning, thereby significantly reducing unplanned downtime, improving production efficiency, extending equipment life, and ensuring stable product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 A schematic diagram of the steps of an abnormal alarm method for feed additive production and processing equipment according to the present invention;

[0081] Figure 2 The present invention is a schematic diagram of the flow chart of an abnormal alarm system for feed additive production and processing equipment. DETAILED DESCRIPTION

[0082] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0083] Example 1

[0084] See also Figure 1 The present invention provides a method for abnormal alarm of feed additive production and processing equipment, comprising the following steps:

[0085] Step 1: Divide the feed additive production area into several production areas according to the production process; the production area includes raw material processing area, mixing area, granulation area, drying area and packaging area;

[0086] Step 2: Set up several monitoring points in each area; install image acquisition equipment and sensor equipment at each monitoring point, collect and obtain the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S of the production equipment at several monitoring points in real time, and establish an abnormal monitoring data set;

[0087] Step 3: After filtering, denoising and format conversion of the abnormal monitoring data set, the parameter range of the production equipment under normal operating conditions is analyzed based on the historical data set, and the alarm threshold of each parameter is set;

[0088] The operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operation time S of the production equipment are compared with the preset alarm thresholds. When the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operation time S of the production equipment exceed the corresponding alarm threshold range, the corresponding warning information is triggered;

[0089] Step 4: Use the image acquisition device installed at each monitoring point to take real-time images of the stirring rod, cutting tool or roller surface in the production equipment to obtain a second image data set; after processing the second image data set, establish a defect recognition model, use an image processing algorithm to perform edge monitoring and morphological processing on the second image data set, and then extract the shape, area and position of the adhesion, evaluate the impact of the adhesion on the equipment operation efficiency, and construct the adhesion impact factor A for each production equipment. yx ;

[0090] Step 5: Adhesion influence factor A for each production equipment yx An assessment is performed to obtain a second assessment result, and corresponding warning information is triggered according to the second assessment result.

[0091] In this embodiment, by subdividing the production area and setting multiple monitoring points, combined with real-time data collection of image acquisition and sensor equipment, the present invention can comprehensively monitor multiple key parameters of equipment operation, including but not limited to temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S. This comprehensive monitoring system can capture subtle changes in the operating status of the equipment, especially those early signals that indicate potential failures, thereby achieving intelligent early warning, preventing equipment failures in advance, and ensuring production continuity and product quality. The present invention analyzes the parameter range of production equipment under normal operating conditions through historical data, and sets the alarm threshold of each parameter accordingly. This dynamic threshold setting mechanism can self-adjust according to the actual operating conditions and historical performance of the equipment, avoiding the false alarm or missed alarm problem caused by traditional fixed thresholds, and improving the accuracy and reliability of the alarm system. In response to the problem of adhesions in key parts inside the equipment (such as stirring rods and cutting tools), the present invention can automatically identify and quantify the shape, area and position of adhesions through real-time monitoring of image acquisition equipment and image processing algorithms, and construct an adhesion influence factor A. yx This innovation can not only evaluate the impact of adhesion on equipment operating efficiency, but also intelligently guide maintenance work to ensure targeted and efficient maintenance, avoid the subjectivity and uncertainty of traditional manual maintenance, reduce maintenance costs, and improve production efficiency.

[0092] The present invention can provide data-driven support for equipment maintenance and production decision-making through in-depth analysis of abnormal monitoring data sets. By comparing historical data with current data, the system can identify trend changes in equipment performance, predict potential failure risks, and provide a scientific basis for the formulation of preventive maintenance plans, thereby optimizing production processes, reducing unplanned downtime, and improving overall production efficiency.

[0093] Example 2, this example is an explanation of Example 1, specifically, a number of monitoring points are set in each production area, and arranged according to the key positions of the production equipment;

[0094] Each monitoring point should cover key production equipment and components, including stirring rods, mixing barrel walls, conveying pipelines and granulator molds; the following equipment should be installed at each monitoring point:

[0095] Temperature sensor, used to monitor the operating temperature T of production equipment;

[0096] Noise sensor, used to monitor the noise value Z during the operation of production equipment;

[0097] Liquid level sensor, used to monitor the liquid level height G of liquid raw materials or products;

[0098] Position sensor, used to monitor the position movement offset value P of the production equipment;

[0099] The operating time sensor is used to monitor the cumulative operating time S of the production equipment.

[0100] In this embodiment, by setting monitoring points around key production equipment and components, such as the stirring rod, the wall of the mixing barrel, the conveying pipeline and the vicinity of the pelletizer mold, detailed data on the equipment operation status can be collected in real time, including temperature T, noise value Z, liquid level height G, position movement offset value P and cumulative operation time S. Real-time monitoring of these data enables the system to respond quickly to any abnormal changes in the equipment, issue early warnings in time, prevent small problems from turning into major failures, and ensure the stability and continuity of production.

[0101] Embodiment 3: This embodiment is an explanation of Embodiment 1. Specifically, the step 3 includes:

[0102] S31, collecting historical data, analyzing the historical data, determining the parameter range of the equipment under normal operating conditions, and obtaining analysis results, including: normal operating temperature range, normal noise level, normal liquid level height, normal position offset range, and normal operating time range;

[0103] S32. According to the analysis results, set the initial alarm threshold of each parameter, wherein the initial alarm threshold is within the upper limit of the normal operation of the equipment; set the temperature threshold Noise Threshold Liquid level threshold Motion offset threshold and runtime thresholds

[0104]

[0105] S33, the operating temperature T of the production equipment is compared with the temperature threshold For comparison, when the operating temperature of the production equipment T> the temperature threshold Calculate the first temperature difference Generate a first temperature warning message, indicating that the temperature exceeds the preset temperature threshold range and there is a risk of damage to the production equipment, including overheating of the equipment due to excessive temperature, aging or burning of components;

[0106] The noise value Z and the noise threshold For comparison, when the noise value Z> noise threshold Calculate the first noise difference Generate the first noise warning information, indicating that the equipment has mechanical failure or component wear, and excessive noise indicates bearing damage, gear wear or insufficient lubrication;

[0107] The liquid level height G and the liquid level threshold For comparison, when the liquid level G> liquid level threshold Calculate the first level difference Generate the first liquid level warning information. Too high a liquid level may cause overflow or equipment damage. When the liquid level height G is less than the liquid level threshold Second level difference Generate a second liquid level warning message. If the liquid level is too low, the pump will run dry or the material supply will be insufficient.

[0108] The moving offset value P and the moving offset threshold For comparison, when the moving offset value P> moving offset threshold Calculate the first moving offset difference Generate the first collision warning information, the equipment position deviation leads to alignment error, product quality problems or equipment collision damage;

[0109] The cumulative running time S is compared with the running time threshold For comparison, when the cumulative running time S> running time threshold Calculate the first run time difference The first timeout warning information is generated, indicating that long-term operation will cause increased equipment wear, depletion of lubricating oil, or fatigue failure of components.

[0110] In this embodiment, the establishment of the above early warning mechanism, through temperature warning, noise warning, liquid level warning, position deviation warning and running time warning, enables potential problems to be identified and solved at an early stage, reduces unplanned downtime and equipment maintenance costs, and avoids production delays caused by equipment failure. A non-single early warning method is adopted, that is, a combination of multiple sensors, data analysis technologies and automated response mechanisms. By integrating the automated response function, the non-single early warning method can automatically perform predefined operations when abnormal signals are detected.

[0111] Embodiment 4: This embodiment is an explanation of Embodiment 1. Specifically, the step 4 includes:

[0112] S41. Install high-definition cameras or industrial cameras at each monitoring point to ensure coverage of key parts such as the stirring rod, cutting tool and roller; high-definition images can provide rich details, which helps to more accurately identify the characteristics of the adhesion and improve the accuracy of subsequent analysis;

[0113] S42. Set the image acquisition frequency, including 5 frames per second or 3 frames per second, to acquire images of the surface of the production equipment in real time and construct a second image data set; according to the operating speed of the production equipment and the accumulation rate of adhesions, reasonably set the image acquisition frequency to ensure that changes in adhesions can be captured in a timely manner without causing excessive burden on data processing.

[0114] S43, filtering and denoising the second image data set, adjusting image contrast and brightness, and performing enhancement processing;

[0115] S44, use the Canny edge detection algorithm to identify the edges in the image, use morphological operations, including dilation, erosion, opening and closing operations to process the edge image, further separate the adhesion and background, and remove noise and small artifacts;

[0116] S45, identifying the contour of the adhesion by a contour monitoring algorithm, extracting the shape of the adhesion from the image, calculating the area of ​​the adhesion according to the contour, and determining the position of the adhesion in the image, and representing the position by a bounding box or an outer rectangle;

[0117] S46. Collect and annotate sample images, mark the shape, area, density and position features of different types of adhesions, construct training sets and test sets, use machine learning or convolutional neural network CNN to train the annotated sample images, establish a defect recognition model that can automatically identify and classify different types of adhesions, and use the second image data set as input to the defect recognition model to calculate and analyze: the adhesion influence factor A of each production equipment yx .

[0118] Train and test the defect recognition model, build and train a machine learning model that can automatically identify and classify adhesions, and use deep learning technologies such as CNN to improve recognition accuracy and generalization capabilities. Automatic recognition and classification of adhesions not only improves efficiency, but also reduces subjective errors caused by manual intervention, and realizes intelligent and standardized management of adhesions.

[0119] According to factors such as the type, area, density and location of adhesions, a quantitative indicator reflecting the degree of influence of adhesions on the performance of production equipment is calculated - the adhesion influence factor A. yx Adhesion influence factor A yx It can intuitively reflect the impact of adhesion on production efficiency, energy consumption and equipment life, help enterprises take timely measures, optimize production and maintenance plans, and reduce unnecessary costs and risks.

[0120] Embodiment 4: This embodiment is an explanation of Embodiment 4. Specifically, S46 includes:

[0121] S461, first calculate the relative adhesion area A: Among them, A total is the total area of ​​the adhesion, calculated by the image processing algorithm, A surface It is the total surface area of ​​the device; the relative adhesion area takes into account the proportion of the adhesion to the total surface area of ​​the device, which can intuitively reflect the coverage rate of the adhesion and is crucial for evaluating the direct impact of the adhesion on the heat dissipation, movement resistance and other aspects of the device.

[0122] S462. Calculate adhesion density Nmd : N is the number of adhesions, which is obtained through image processing algorithms. The adhesion density index reflects the density of adhesions through the ratio of the number of adhesions to the equipment area, which is meaningful for predicting the frequency and intensity of equipment cleaning and maintenance.

[0123] S463, extract and calculate the shape factor C, comprehensively consider the circularity and aspect ratio parameters of the adhesion, and calculate the shape factor C:

[0124]

[0125] In the formula, A i is the area of ​​the ith adhesion, P i is the perimeter of the ith adhesion, L i and W i are the length and width of the ith adhesion, respectively; the shape factor C can reveal the influence of adhesions on the surface friction and fluid dynamics of the equipment by analyzing the geometric properties of adhesions, such as circularity and aspect ratio, and is necessary for evaluating the influence of adhesions on the operating efficiency of the equipment.

[0126] S464. Comprehensively consider the distance from the center of the adhesion to the key components of the equipment and obtain the position factor D:

[0127]

[0128] Where, d i is the distance from the ith adhesion to the key part of the equipment, d max It is the maximum distance from any point on the surface of the equipment to the key parts. The position factor D takes into account the spatial relationship of the adhesion relative to the key parts of the equipment. It is particularly important for identifying the impact of adhesion on sensitive parts of the equipment (such as drive shafts, bearings, etc.), and helps to accurately locate the key points of maintenance.

[0129] Adhesion influence factor A for each production equipment yx Calculated by the following steps:

[0130]

[0131] Where w1, w2, w3 and w4 represent the relative adhesion area A, adhesion density N, respectively. md , the weight values ​​of shape factor C and position factor D, and 0≤w1≤1, 0≤w2≤1, 0≤w3≤1, 0≤w4≤1, and w1+w2+w3+w4=1, d1 is represented as a correction constant.

[0132] In this embodiment, the adhesion influencing factor A yxThe calculation provides a scientific basis for maintenance decisions and avoids over-maintenance or under-maintenance. By quantifying the impact of adhesions, enterprises can optimize production processes and equipment design in a targeted manner to reduce the generation of adhesions, thereby improving production efficiency and equipment operation stability. Accurately evaluating the impact of adhesions helps reduce unnecessary maintenance costs and avoid production interruptions due to equipment failures, thereby achieving effective cost control. Timely cleaning and effective management of adhesions help maintain the safe operation of equipment.

[0133] Example 5: This example is an explanation of Example 1. Specifically, the adhesion influence factor A of each production equipment is yx Conduct an assessment and obtain the second assessment results, including:

[0134] Setting a first impact threshold X1 and a second impact threshold X2, wherein the first impact threshold X1>the second impact threshold X2;

[0135] When the adhesion factor A yx < the second impact threshold X2, indicating that the adhesion state of the production equipment is normal and no warning is triggered;

[0136] When the second impact threshold X2 ≤ adhesion impact factor A yx ≤ the first impact threshold X1, indicating that the adhesion state of the production equipment is abnormal, triggering the first adhesion warning information, which is: "The equipment has a slight adhesion problem, please check and clean it";

[0137] When the adhesion factor A yx >The first impact threshold X1 indicates that the adhesion state of the production equipment is abnormal, triggering the second adhesion warning information. The second adhesion warning information is: "The equipment has serious adhesion problems, please check and clean it."

[0138] In this embodiment, when the adhesion influencing factor A yx When the first impact threshold X1 is between the second impact threshold X2 and the first impact threshold X1, the first adhesion warning message is triggered, indicating that the equipment has a slight adhesion problem, encouraging timely inspection and cleaning. This early warning helps prevent equipment performance degradation or failure caused by adhesion accumulation, reducing the risk of production interruption. yxWhen the first impact threshold X1 is exceeded, the second adhesion warning information is triggered, indicating that the equipment is facing a serious adhesion problem and immediate action is required. This immediate warning mechanism ensures that the enterprise can respond quickly, avoiding equipment damage, a significant drop in production efficiency, and even safety accidents, and effectively controlling potential major economic losses. The implementation of the multi-level warning mechanism makes the formulation of maintenance plans more scientific and reasonable. Enterprises can adjust maintenance strategies according to the warning level to avoid unnecessary excessive maintenance, while also ensuring that timely measures are taken before the adhesion problem worsens, thereby finding the best balance between maintenance costs and equipment performance.

[0139] Example 5. This example is an explanation of Example 1. Specifically, the first temperature difference T1, the first noise difference Z1, the first liquid level difference G1, the second liquid level difference G2, the first movement offset difference P1, and the first operating time difference S1 are dimensionlessly processed, and are prioritized from high to low according to the size of the values. Alarm processing is performed according to the corresponding first temperature warning information, first noise warning information, first liquid level warning information, second liquid level warning information, first collision warning information, and first timeout warning information according to the corresponding priority ranking.

[0140] In this embodiment, the processed differences are sorted according to the numerical values, which can automatically identify the abnormal situations that have the most urgent impact on the operation of the equipment, ensuring that urgent problems are handled with priority. This intelligent sorting mechanism helps to optimize resource allocation, ensuring that within limited maintenance capabilities and time, the abnormalities that are most likely to cause production interruptions or equipment damage are resolved with priority, thereby improving the efficiency and pertinence of maintenance responses. Prioritizing abnormalities that have the greatest impact on safety, such as excessive temperatures, excessive noise, or equipment collision warnings, can effectively prevent potential safety accidents and protect the safety of equipment and operators. This safety-oriented warning processing method not only reduces the possibility of accidents, but also improves the company's level of safety production management.

[0141] Example 6: This example is an explanation of Example 1. Specifically, a first strategy is generated according to the first temperature warning information: immediately and automatically reduce the operating power of the production equipment by 30% or start the cooling system, and send an alarm to the maintenance team for on-site inspection to ensure that the heat dissipation equipment operates normally, and record the temperature abnormality history;

[0142] The second strategy is generated based on the first noise warning information: stop the machine for inspection, check the bearings and gear parts, replace the damaged parts and readjust the lubrication system when the bearings and gear parts are damaged; stop the machine for inspection of the bearings and gear parts, replace the damaged parts in time and adjust the lubrication system, which can not only quickly eliminate the noise source, but also avoid potential mechanical failures, reduce unplanned downtime, and ensure production continuity. This also helps to improve the overall operating efficiency and safety of the equipment;

[0143] The third strategy is generated based on the first liquid level warning information: automatically adjust the feed rate or discharge excess liquid to prevent overflow; automatically adjust the feed rate or discharge excess liquid to effectively prevent material overflow or shortage and maintain material balance in the production process. This not only reduces waste, but also avoids production stagnation caused by improper material management and ensures the smooth operation of the production line.

[0144] A fourth strategy is generated based on the second liquid level warning information: starting a backup pump or manually increasing the supply to prevent production interruptions due to insufficient supply; this helps to improve production efficiency and avoid production delays caused by material supply problems.

[0145] The fifth strategy is generated based on the first collision warning information: emergency stop of related equipment, use of positioning system to correct equipment position, check and adjust guidance device or alignment mechanism, and restart only after confirmation; emergency stop of equipment and position correction can avoid further equipment damage or production accidents, and ensure the safety of operators. Checking and adjusting guidance device or alignment mechanism can help restore normal operation of equipment, reduce unplanned downtime, and maintain production order.

[0146] The sixth strategy is generated based on the first timeout warning information: planned downtime maintenance, performing regular maintenance such as changing lubricants, checking fasteners and wearing parts; it can prevent problems caused by equipment aging and wear, extend equipment life, and reduce the incidence of sudden failures. This helps maintain production efficiency, reduce maintenance costs, and improve the overall reliability and stability of the equipment.

[0147] The seventh strategy is generated based on the first adhesion warning: the cleaning procedure is activated, including the use of an automatic cleaning system or manual cleaning; this avoids equipment efficiency loss or failure due to adhesion. This helps maintain the sanitary conditions of the production line, improves product quality, and also reduces production interruptions caused by adhesion problems.

[0148] The eighth strategy is generated based on the second adhesion warning: full shutdown for cleaning and analysis of the adhesion cause. If it is a problem with the raw material formula, process parameters or equipment design, adjustments must be made before restarting production. Whether it is adjusting the raw material formula, process parameters or improving equipment design, it can significantly improve production efficiency and product quality, while reducing maintenance costs and the risk of production interruptions.

[0149] Example 7, a feed additive production and processing equipment abnormal alarm system, please refer to Figure 2 ,include,

[0150] The area division unit is used to divide the feed additive production area into several production areas according to the production process; the production area includes the raw material processing area, mixing area, granulation area, drying area and packaging area;

[0151] The first monitoring unit is used to install image acquisition equipment and sensor equipment at each monitoring point, collect and obtain the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S of the production equipment at several monitoring points in real time, and establish an abnormal monitoring data set;

[0152] The first processing unit performs data filtering, denoising and format conversion on the abnormal monitoring data set, analyzes the parameter range of the production equipment under normal operating conditions based on the historical data set, and sets the alarm threshold of each parameter;

[0153] The first evaluation unit compares the operating temperature T, noise value Z, liquid level height G, mobile offset value P and cumulative operating time S of the production equipment with the preset alarm threshold. When the operating temperature T, noise value Z, liquid level height G, mobile offset value P and cumulative operating time S of the production equipment exceed the corresponding alarm threshold range, the corresponding warning information is triggered.

[0154] The image acquisition unit uses an image acquisition device installed at each monitoring point to capture the device image of the stirring rod, cutting tool or drum surface in the production equipment in real time to obtain a second image data set;

[0155] Establish a defect recognition model analysis unit, establish a defect recognition model, use image processing algorithms to perform edge monitoring and morphological processing on the second image data set, extract the shape, area and position of the adhesion, evaluate the impact of adhesion on equipment operation efficiency, and construct the adhesion impact factor A for each production equipment yx ;

[0156] The second evaluation unit is used to evaluate the adhesion influence factor A of each production equipment. yx Performing an assessment to obtain a second assessment result, and triggering corresponding warning information according to the second assessment result;

[0157] The priority unit performs dimensionless processing on the first temperature difference T1, the first noise difference Z1, the first liquid level difference G1, the second liquid level difference G2, the first movement offset difference P1, and the first operation time difference S1, and prioritizes them from high to low according to the values, and performs corresponding priority sorting and alarm processing on the corresponding warning information;

[0158] The strategy generating unit generates a corresponding strategy according to the first temperature warning information, the first noise warning information, the first liquid level warning information, the second liquid level warning information, the first collision warning information, the first timeout warning information, the first adhesion warning information and the second adhesion warning information.

[0159] In this embodiment, the system realizes real-time and accurate control of the operating status of production equipment by finely dividing and comprehensively monitoring the production area. The system integrates the monitoring and analysis of multi-dimensional data such as temperature, noise, liquid level, movement deviation and running time, and combines image acquisition and defect recognition technology. It can not only immediately detect and warn of equipment abnormalities such as overheating, excessive noise, abnormal liquid level, movement deviation and adhesion problems, but also prioritize according to the severity of the warning information to ensure that key issues are given priority. The system improves the accuracy and effectiveness of warnings by dynamically setting alarm thresholds and using historical data to analyze the normal operating parameter range of equipment. In particular, the analysis of equipment images through image processing algorithms can quantitatively evaluate the impact of adhesions on equipment operating efficiency, construct adhesion impact factors, and provide an intuitive basis for equipment maintenance. In addition, the emergency strategies automatically generated by the system can guide the maintenance team to respond quickly and take corresponding measures, such as adjusting power, starting the cooling system, stopping for inspection, automatically adjusting feed or discharge, starting standby pumps, emergency stopping to correct equipment position, performing regular maintenance or comprehensive cleaning, thereby significantly reducing unplanned downtime, improving production efficiency, extending equipment life, and ensuring stable product quality.

[0160] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for abnormal alarm of feed additive production and processing equipment, characterized in that: The following steps are included: Step 1: Divide the feed additive production area into several production areas according to the production process; the production area includes raw material processing area, mixing area, granulation area, drying area and packaging area; Step 2: Set up several monitoring points in each area; install image acquisition equipment and sensor equipment at each monitoring point, collect and obtain the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S of the production equipment at several monitoring points in real time, and establish an abnormal monitoring data set; Step 3: After filtering, denoising and format conversion of the abnormal monitoring data set, the parameter range of the production equipment under normal operating conditions is analyzed based on the historical data set, and the alarm threshold of each parameter is set; The operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operation time S of the production equipment are compared with the preset alarm thresholds. When the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operation time S of the production equipment exceed the corresponding alarm threshold range, the corresponding warning information is triggered; Step 4: Use the image acquisition device installed at each monitoring point to capture the equipment image of the stirring rod, cutting tool or roller surface in the production equipment in real time to obtain the second image data set; after processing the second image data set, establish a defect recognition model, use the image processing algorithm to perform edge monitoring and morphological processing on the second image data set, and then extract the shape, area and position of the adhesion, evaluate the impact of the adhesion on the equipment operation efficiency, and construct the adhesion impact factor A of each production equipment. yx ; The fourth step comprises: S41. Install high-definition cameras or industrial cameras at each monitoring point to ensure coverage of key parts of the stirring rod, cutting tool and drum; S42, setting the image acquisition frequency, including 5 frames per second or 3 frames per second, to acquire images of the surface of the production equipment in real time and construct a second image data set; S43, filtering and denoising the second image data set, adjusting image contrast and brightness, and performing enhancement processing; S44, use the Canny edge detection algorithm to identify the edges in the image, use morphological operations, including dilation, erosion, opening and closing operations to process the edge image, further separate the adhesion and background, and remove noise and small artifacts; S45, identifying the contour of the adhesion by a contour monitoring algorithm, extracting the shape of the adhesion from the image, calculating the area of ​​the adhesion according to the contour, and determining the position of the adhesion in the image, and representing the position by a bounding box or an outer rectangle; S46. Collect and annotate sample images, mark the shape, area, density and position features of different types of adhesions, construct training sets and test sets, use machine learning or convolutional neural network CNN to train the annotated sample images, establish a defect recognition model that can automatically identify and classify different types of adhesions, and use the second image data set as input to the defect recognition model to calculate and analyze: the adhesion influence factor A of each production equipment yx ; The S46 includes: S461, first calculate the relative adhesion area A: Among them, A total is the total area of ​​the adhesion, calculated by the image processing algorithm, A surface is the total surface area of ​​the equipment; S462. Calculate adhesion density N md : N is the number of adhesions, which is obtained by counting the image processing algorithm; S463, extract and calculate the shape factor C, comprehensively consider the circularity and aspect ratio parameters of the adhesion, and calculate the shape factor C: In the formula, A i is the area of ​​the ith adhesion, P i is the perimeter of the ith adhesion, L i and W i are the length and width of the ith adhesion, respectively; S464. Comprehensively consider the distance from the center of the adhesion to the key components of the equipment and obtain the position factor D: Where, d i is the distance from the ith adhesion to the key part of the equipment, d max It is the maximum distance from any point on the surface of the equipment to the key part; Adhesion influence factor A for each production equipment yx Calculated by the following steps: Where w1, w2, w3 and w4 represent the relative adhesion area A, adhesion density N, respectively. md , the weight values ​​of the shape factor C and the position factor D, and 0≤w1≤1, 0≤w2≤1, 0≤w3≤1, 0≤w4≤1, and w1+w2+w3+w4=1, d1 is represented as the correction constant; Step 5, for each production equipment adhesion influence factor A yx An assessment is performed to obtain a second assessment result, and corresponding warning information is triggered according to the second assessment result.

2. The method for abnormal alarm of feed additive production and processing equipment according to claim 1, characterized in that: Set up several monitoring points in each production area and arrange them according to the key locations of production equipment; Each monitoring point should cover key production equipment and components, including stirring rods, mixing barrel walls, conveying pipelines and granulator molds; the following equipment should be installed at each monitoring point: Temperature sensor, used to monitor the operating temperature T of production equipment; Noise sensor, used to monitor the noise value Z during the operation of production equipment; Liquid level sensor, used to monitor the liquid level height G of liquid raw materials or products; Position sensor, used to monitor the position movement offset value P of the production equipment; The operating time sensor is used to monitor the accumulated operating time S of the production equipment.

3. The method for abnormal alarm of feed additive production and processing equipment according to claim 1, characterized in that: The step three comprises: S31, collecting historical data, analyzing the historical data, determining the parameter range of the equipment under normal operating conditions, and obtaining analysis results, including: normal operating temperature range, normal noise level, normal liquid level height, normal position offset range, and normal operating time range; S32. According to the analysis results, set the initial alarm threshold of each parameter, wherein the initial alarm threshold is within the upper limit of the normal operation of the equipment; set the temperature threshold Noise Threshold Liquid level threshold Motion offset threshold and runtime thresholds S33, the operating temperature T of the production equipment is compared with the temperature threshold For comparison, when the operating temperature of the production equipment T> the temperature threshold Calculate the first temperature difference Generate a first temperature warning message, indicating that the temperature exceeds the preset temperature threshold range and there is a risk of damage to the production equipment, including overheating of the equipment due to excessive temperature, aging or burning of components; The noise value Z and the noise threshold For comparison, when the noise value Z> noise threshold Calculate the first noise difference Generate the first noise warning information, indicating that the equipment has mechanical failure or component wear, and excessive noise indicates bearing damage, gear wear or insufficient lubrication; The liquid level height G and the liquid level threshold For comparison, when the liquid level G> liquid level threshold Calculate the first level difference Generate the first liquid level warning information. Too high a liquid level may cause overflow or equipment damage. When the liquid level height G is less than the liquid level threshold Second level difference Generate a second liquid level warning message. If the liquid level is too low, the pump will run dry or the material supply will be insufficient. The moving offset value P and the moving offset threshold For comparison, when the moving offset value P> moving offset threshold Calculate the first moving offset difference Generate the first collision warning information, the equipment position deviation leads to alignment error, product quality problems or equipment collision damage; The cumulative running time S is compared with the running time threshold For comparison, when the cumulative running time S> running time threshold Calculate the first run time difference The first timeout warning information is generated, indicating that long-term operation will cause increased equipment wear, depletion of lubricating oil, or fatigue failure of components.

4. The method for abnormal alarm of feed additive production and processing equipment according to claim 1, characterized in that: Adhesion influence factor A for each production equipment yx Conduct an assessment and obtain the second assessment results, including: Setting a first impact threshold X1 and a second impact threshold X2, wherein the first impact threshold X1>the second impact threshold X2; When the adhesion factor A yx < the second impact threshold X2, indicating that the adhesion state of the production equipment is normal and no warning is triggered; When the second impact threshold X2 ≤ adhesion impact factor A yx ≤ the first impact threshold X1, indicating that the adhesion state of the production equipment is abnormal, triggering the first adhesion warning information, which is: "The equipment has a slight adhesion problem, please check and clean it"; When the adhesion factor A yx >The first impact threshold X1 indicates that the adhesion state of the production equipment is abnormal, triggering the second adhesion warning information. The second adhesion warning information is: "The equipment has serious adhesion problems, please check and clean it." 5. The method for abnormal alarm of feed additive production and processing equipment according to claim 1, characterized in that: The first temperature difference T1, the first noise difference Z1, the first liquid level difference G1, the second liquid level difference G2, the first movement offset difference P1, and the first operating time difference S1 are dimensionlessly processed and prioritized from high to low according to the magnitude of the values, and alarm processing is performed according to the corresponding first temperature warning information, first noise warning information, first liquid level warning information, second liquid level warning information, first collision warning information, and first timeout warning information.

6. The method for abnormal alarm of feed additive production and processing equipment according to claim 5, characterized in that: Generate the first strategy based on the first temperature warning information: immediately and automatically reduce the operating power of the production equipment by 30% or start the cooling system, and send an alarm to the maintenance team for on-site inspection to ensure that the cooling equipment is operating normally and record the temperature abnormality history; Generate a second strategy based on the first noise warning information: stop the machine for inspection, check the bearings and gear components, and replace the damaged components and readjust the lubrication system when the bearings and gear components are damaged; Generate a third strategy based on the first liquid level warning information: automatically adjust the feed rate or discharge excess liquid to prevent overflow; Generate a fourth strategy based on the second liquid level warning information: start the standby pump or manually increase the supply to prevent production interruption due to insufficient supply; Generate a fifth strategy based on the first collision warning information: emergency stop the relevant equipment, use the positioning system to correct the position of the equipment, check and adjust the guidance device or alignment mechanism, and restart after confirmation; Generate the sixth strategy based on the first timeout warning information: planned shutdown maintenance, perform regular maintenance such as changing lubricating oil, checking fasteners and wearing parts; Generate the seventh strategy based on the first adhesion warning: initiate cleaning procedures, including using an automatic cleaning system or arranging manual cleaning; The eighth strategy is generated based on the second adhesion warning: shut down the entire machine for cleaning and analyze the cause of adhesion. If it is a problem with the raw material formula, process parameters or equipment design, adjustments must be made before restarting production.

7. A feed additive production and processing equipment abnormality alarm system, applied to a feed additive production and processing equipment abnormality alarm method according to any one of claims 1 to 6, characterized in that: include, The area division unit is used to divide the feed additive production area into several production areas according to the production process; the production area includes the raw material processing area, mixing area, granulation area, drying area and packaging area; The first monitoring unit is used to install image acquisition equipment and sensor equipment at each monitoring point, collect and obtain the operating temperature T, noise value Z, liquid level height G, movement offset value P and cumulative operating time S of the production equipment at several monitoring points in real time, and establish an abnormal monitoring data set; The first processing unit performs data filtering, denoising and format conversion on the abnormal monitoring data set, analyzes the parameter range of the production equipment under normal operating conditions based on the historical data set, and sets the alarm threshold of each parameter; The first evaluation unit compares the operating temperature T, noise value Z, liquid level height G, mobile offset value P and cumulative operating time S of the production equipment with the preset alarm threshold. When the operating temperature T, noise value Z, liquid level height G, mobile offset value P and cumulative operating time S of the production equipment exceed the corresponding alarm threshold range, the corresponding warning information is triggered. The image acquisition unit uses an image acquisition device installed at each monitoring point to capture the device image of the stirring rod, cutting tool or drum surface in the production equipment in real time to obtain a second image data set; Establish a defect recognition model analysis unit, establish a defect recognition model, use image processing algorithms to perform edge monitoring and morphological processing on the second image data set, extract the shape, area and position of the adhesion, evaluate the impact of adhesion on equipment operation efficiency, and construct the adhesion impact factor A for each production equipment yx ; The second evaluation unit is used to evaluate the adhesion influence factor A of each production equipment. yx Performing an assessment to obtain a second assessment result, and triggering corresponding warning information according to the second assessment result; The priority unit performs dimensionless processing on the first temperature difference T1, the first noise difference Z1, the first liquid level difference G1, the second liquid level difference G2, the first movement offset difference P1, and the first operation time difference S1, and prioritizes them from high to low according to the values, and performs corresponding priority sorting and alarm processing on the corresponding warning information; The strategy generating unit generates a corresponding strategy according to the first temperature warning information, the first noise warning information, the first liquid level warning information, the second liquid level warning information, the first collision warning information, the first timeout warning information, the first adhesion warning information and the second adhesion warning information.

Images