A feed quality real-time regulation method and system based on multi-modal data

Abstract

The present application relates to the technical field of device control, and in particular to a feed quality real-time regulation method and system based on multi-modal data, which mainly includes constructing a feed quality evaluation model, determining whether the current feed is normal based on a quality index, outputting a device regulation index based on the frequency and the deviation rate of the data corresponding to the current target device through a device regulation model, sorting the device regulation index, and outputting the regulation. The above method gives a relatively objective sorting result, and the control system or the staff can adjust the parameters of the corresponding device according to the current sorting result, can complete the regulation in the shortest and most reasonable time, realize the continuous production of the production line, improve the detection efficiency and realize the relatively effective regulation, and at the same time, avoid the traditional static regulation mode, and adjust the current feed production in real time.

Description

Technical Field

[0001] This invention relates to the field of equipment control technology, and more specifically, to a method and system for real-time control of feed quality based on multimodal data. Background Technology

[0002] As the core foundation of the livestock industry, feed quality directly determines breeding efficiency, animal health, and the safety of livestock products. On the production side, relying on production equipment, processes such as crushing, mixing, and pelleting are carried out. By adjusting mixing time, temperature, and moisture, the uniformity of feed mixing and processing quality are ensured. Simultaneously, a digital management platform is built, integrating raw material testing data, production parameters, and breeding feedback information, based on the nutritional needs of animals at different growth stages. Regular sampling and testing of finished feed verifies compliance with quality standards.

[0003] However, this traditional static control mode is difficult to adapt to the dynamic changes in raw materials and the differentiated nutritional needs of animals. It cannot avoid safety risks such as mycotoxins and impurities in a timely manner. Secondly, when quality problems occur, the relevant equipment is usually adjusted based on experience, but this is inefficient and it is unclear which equipment should be adjusted first to achieve better results. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for real-time control of feed quality based on multimodal data, so as to solve the above-mentioned problems in the prior art.

[0005] This invention is achieved through the following technical solution:

[0006] In a first aspect, the present invention provides a method for real-time control of feed quality based on multimodal data, comprising:

[0007] Acquire multimodal data of the current feed, calculate the deviation rate of each data in the multimodal data, construct a feed quality evaluation model, output the quality index of the current feed based on the deviation rate of each data, and determine whether the current feed is normal based on the quality index;

[0008] If it is normal, no action is taken; if it is abnormal, a deviation threshold is set, and data with a deviation rate greater than the deviation threshold is taken as target abnormal data.

[0009] Construct a data-device relationship mapping table, obtain the target device corresponding to each type of target abnormal data, obtain the frequency of occurrence of each target device, construct a device control model, and output the device control index based on the frequency and the deviation rate of the data corresponding to the current target device through the device control model;

[0010] The control indices of the devices are sorted. A control signal is sent to the target device ranked first. After receiving a feedback signal that the target device ranked first has completed the control, the quality index is acquired again. If the quality index is normal, the control is stopped. If it is not normal, the control signal is sent sequentially to the next target device until the quality index is normal. If the quality index is still not normal after all target devices have been controlled, an alarm signal is issued and a control signal to stop all devices is output.

[0011] Preferably, acquiring the multimodal data of the current feed includes:

[0012] Determine whether the types of multimodal data to be detected match the preset detection types. If yes, collect data normally. If no, obtain the currently missing types and collect the data corresponding to the currently missing types again.

[0013] If data corresponding to the currently missing type is collected, collection proceeds normally; if data corresponding to the currently missing type is still not collected, an early warning signal is sent to the control terminal.

[0014] Preferably, the multimodal data includes particle size uniformity index, mixing uniformity, moisture content, and particle integrity, wherein the particle size uniformity index includes:

[0015]

[0016] In the formula, The particle size uniformity index is... For the first Sample quality on the sieve For the first The average pore size on the sieve, It represents the geometric mean particle size.

[0017] Preferably, the particle integrity includes:

[0018] Acquire the image data of the current feed, construct an image recognition model, identify the feed particles in the current image data using the image recognition model, and output the outline of the feed particles;

[0019] Extract the target contour image of each feed pellet, compare it with the standard contour image of a standard feed pellet to obtain the similarity, set a similarity threshold, and output complete feed pellets and incomplete feed pellets based on the similarity and similarity threshold. The pellet integrity is obtained based on the number of complete and incomplete feed pellets.

[0020] Preferably, the construction of the image recognition model includes:

[0021] After adaptive grayscale normalization of the image data, bilateral filtering is performed to obtain the image to be identified;

[0022] The system captures the edges of particles of different sizes in the image to be identified, sets the condition for judging points on the edges as contour points, obtains several contour points, smooths and completes these contour points, and outputs the complete contour.

[0023] Preferably, the step of performing bilateral filtering to obtain the image to be identified includes:

[0024]

[0025]

[0026]

[0027]

[0028] In the formula, The pixels of the image to be identified after bilateral filtering grayscale value, For pixels grayscale value at that location Spatial weights, For grayscale weights, For direction weights, For space core, For grayscale kernel, For pixels With pixels The angle between the line connecting the particles and the main direction of the particle.

[0029] Preferably, the step of capturing the edges of particles of different sizes in the image to be identified and setting the conditions for determining points on the edges as contour points includes:

[0030] Perform multi-scale gradient calculation:

[0031]

[0032] The condition for a point to be a contour point is to satisfy the following conditions:

[0033]

[0034]

[0035] In the formula, To fuse the magnitude of the gradient, For scale weights, For the scale is Gaussian smoothed image gradient, Let be the standard deviation of the k-th Gaussian smoothed scale. To increase the gradient threshold, The global gradient mean. The average gray level of the background. The grayscale difference threshold. The standard deviation of the background grayscale. For pixels The grayscale value at that location.

[0036] Preferably, the construction of the feed quality evaluation model includes:

[0037]

[0038] In the formula, As an evaluation index, The standard particle size uniformity index. For standard mixing uniformity, This represents the current particle size uniformity index. Given the current moisture content, Standard moisture content, The total number of particles, The total number of complete particles, , , and The weights are calculated, and their sum is 1.

[0039] Preferably, the constructed device control model includes:

[0040]

[0041] In the formula, For the first Equipment control index of each device. The frequency of the current device's occurrence. This represents the number of target abnormal data associated with the current device. This represents the number of normal data items currently associated with the device. The adjustment time required for the current equipment. For the first The deviation rate of the target abnormal data.

[0042] Secondly, the present invention also provides a real-time feed quality control system based on multimodal data, used to execute the above-mentioned real-time feed quality control method based on multimodal data, comprising:

[0043] The data processing module is configured to acquire multimodal data of the current feed, calculate the deviation rate of each data type in the multimodal data, construct a feed quality evaluation model, output a quality index of the current feed based on the deviation rate of each data type, determine whether the current feed is normal based on the quality index; if normal, no processing is performed; if abnormal, a deviation threshold is set, and data with a deviation rate greater than the deviation threshold are designated as target abnormal data; construct a data-equipment mapping table, acquire the target equipment corresponding to each type of target abnormal data, obtain the frequency of occurrence of each target equipment, construct an equipment control model, and output an equipment control index based on the frequency and the deviation rate of the data corresponding to the current target equipment.

[0044] The execution module is configured to sort the control indices of the devices, send a control signal to the first target device in the sort, and after receiving a feedback signal that the first target device has completed the control, acquire the quality index again. If the quality index is normal, the control is stopped; if it is not normal, the control signals are sent sequentially to the next target device until the quality index is normal. If the quality index is still not normal after all target devices have been controlled, an alarm signal is issued and a control signal to stop all devices is output.

[0045] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0046] The method provided by this invention mainly includes constructing a feed quality evaluation model, judging whether the current feed is normal based on a quality index, outputting an equipment control index through an equipment control model based on the frequency and deviation rate of the data corresponding to the current target equipment, ranking the equipment control indices, and outputting control. This method provides a relatively objective ranking result, allowing the control system or operators to adjust the parameters of the corresponding equipment based on the current ranking result. This enables complete control within a short and reasonable timeframe, ensuring continued production on the production line, improving detection efficiency, and achieving relatively effective control. It also avoids the traditional static control mode, allowing for real-time adjustments to current feed production. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart illustrating the present invention.

[0049] Figure 2This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0051] The module division in this application is a logical division. In actual application, there may be other division methods. For example, multiple modules may be combined into or integrated into another system, or some features may be ignored or not executed.

[0052] The independently described modules or sub-modules may or may not be physically separated; they may be implemented in software or hardware, and some modules or sub-modules may be implemented in software, with the processor calling the software to implement the function of these modules or sub-modules, while other modules or sub-modules may be implemented in hardware, such as through hardware circuits. Furthermore, some or all of the modules can be selected to achieve the purpose of this application's solution according to actual needs.

[0053] Please refer to Figures 1-2 This invention provides a method for real-time control of feed quality based on multimodal data, comprising:

[0054] S101: Take the multimodal data of the current feed, calculate the deviation rate of each data in the multimodal data, construct a feed quality evaluation model, output the quality index of the current feed based on the deviation rate of each data, and determine whether the current feed is normal based on the quality index;

[0055] In this embodiment, the modal testing data selected include particle size uniformity index, mixing uniformity, moisture content, and particle integrity. Mixing uniformity and moisture content can be directly measured by relevant instruments. The deviation rate of this scheme is defined as the absolute value of the difference between the current value and the standard value divided by the standard value.

[0056] S102: If normal, no action is taken; if abnormal, a deviation threshold is set, and data with a deviation rate greater than the deviation threshold is taken as target abnormal data.

[0057] S103: Construct a data-device relationship mapping table, obtain the target device corresponding to each type of target abnormal data, obtain the frequency of occurrence of each target device, construct a device control model, and output the device control index based on the frequency and the deviation rate of the data corresponding to the current target device through the device control model.

[0058] In this embodiment, the relationship mapping table represents the relationship between data and devices. It can be understood as which devices are needed to control the transformation of the current data. For example, the current x-th type of data needs to be controlled by devices A, B, and C, and the y-th type of data needs to be controlled by devices B and D. After listing all the data, a relationship mapping table is formed, which is then extracted during subsequent keyword retrieval.

[0059] S103: Sort the control index of the equipment, send a control signal to the target equipment ranked first, and after receiving the feedback signal that the target equipment ranked first has completed the control, obtain the quality index again. If the quality index is normal, stop the control. If it is not normal, send control signals to the next target equipment in sequence until the quality index is normal. If the quality index is still not normal after all target equipment is controlled, issue an alarm signal and output a control signal to stop all equipment.

[0060] If the abnormality persists even after all the corresponding equipment for the abnormal data has been adjusted, the entire production line needs to be stopped.

[0061] The method provided by this invention mainly includes constructing a feed quality evaluation model, judging whether the current feed is normal based on a quality index, outputting an equipment control index through an equipment control model based on the frequency and deviation rate of the data corresponding to the current target equipment, ranking the equipment control indices, and outputting control. This method provides a relatively objective ranking result, allowing the control system or operators to adjust the parameters of the corresponding equipment based on the current ranking result. This enables complete control within a short and reasonable timeframe, ensuring continued production on the production line, improving detection efficiency, and achieving relatively effective control. It also avoids the traditional static control mode, allowing for real-time adjustments to current feed production.

[0062] In one exemplary embodiment of the present invention, acquiring the multimodal data of the current feed includes:

[0063] Determine whether the type of multimodal data to be detected matches the preset detection type. If yes, collect normally. If no, obtain the currently missing type and collect the data corresponding to the currently missing type again. If the data corresponding to the currently missing type is collected, collect normally. If the data corresponding to the currently missing type is still not collected, issue an early warning signal to the control terminal.

[0064] In this embodiment, it is necessary to ensure the integrity of the data before proceeding with subsequent testing and evaluation.

[0065] In one exemplary embodiment of the present invention, the multimodal data includes particle size uniformity index, mixing uniformity, moisture content, and particle integrity, wherein the particle size uniformity index includes:

[0066]

[0067] In the formula, The particle size uniformity index is... For the first Sample quality on the sieve For the first The average pore size on the sieve, It represents the geometric mean particle size.

[0068] Specifically, the particle integrity includes:

[0069] Acquire the image data of the current feed, construct an image recognition model, identify the feed particles in the current image data using the image recognition model, and output the outline of the feed particles;

[0070] Extract the target contour image of each feed pellet, compare it with the standard contour image of a standard feed pellet to obtain the similarity, set a similarity threshold, and output complete feed pellets and incomplete feed pellets based on the similarity and similarity threshold. The pellet integrity is obtained based on the number of complete and incomplete feed pellets.

[0071] It should be noted that when collecting image data of feed, all pellets should be laid out in one layer to avoid overlapping. Overlapping pellets will also show defects, which will lead to large errors in subsequent timing. Therefore, it is sufficient to collect images of pellets laid out in one layer. On the other hand, the integrity of the pellets reflects whether there are many defects in the pellets during operation due to insufficient viscosity.

[0072] Regarding similarity, a coordinate system is established with the geometric center of the contour in the geometric contour image as the origin. The coordinates of the contour pixels in the standard contour image and the target contour image are obtained respectively. Starting from the first pixel coordinate at the same position, the distance difference between the pixel coordinates at the same position in the two images is calculated, and a judgment threshold is set. If the distance difference is greater than the judgment threshold, it is considered that the contour pixels of the current target contour image deviate from the contour pixels of the standard contour image. All pixels are judged to be deviated. The similarity is obtained by the ratio of the number of deviated pixels to the total number of pixels. If the similarity is greater than the similarity threshold, it is considered to be a complete feed pellet; otherwise, it is considered incomplete.

[0073]

[0074] In the formula, This is the distance difference. , These are the coordinates of the contour pixels in a standard contour image. , The coordinates of the contour pixels of the target contour image.

[0075] The first pixel coordinate can be the pixel with a ordinate of 0 that is closest to the origin to the left of the origin.

[0076] The image recognition model is constructed by: performing adaptive grayscale normalization on the image data, then performing bilateral filtering to obtain the image to be recognized; capturing the edges of particles of different sizes in the image to be recognized, and setting the condition for judging points on the edges as contour points to obtain several contour points; smoothing and completing the contour points to output the complete contour.

[0077] Specifically, the image to be identified by bilateral filtering includes:

[0078]

[0079]

[0080]

[0081]

[0082] In the formula, The pixels of the image to be identified after bilateral filtering grayscale value, For pixels grayscale value at that location Spatial weights, For grayscale weights, For direction weights, For space core, For grayscale kernel, For pixels With pixels The angle between the line connecting the particles and the main direction of the particle.

[0083] The process involves capturing the edges of particles of different sizes in the image to be identified, and setting conditions for determining points on the edges as contour points, including:

[0084] Perform multi-scale gradient calculation:

[0085]

[0086] The condition for a point to be a contour point is to satisfy the following conditions:

[0087]

[0088]

[0089] In the formula, To fuse the magnitude of the gradient, For scale weights, For the scale is Gaussian smoothed image gradient, Let be the standard deviation of the k-th Gaussian smoothed scale. To increase the gradient threshold, The global gradient mean. The average gray level of the background. The grayscale difference threshold. The standard deviation of the background grayscale. For pixels The grayscale value at that location.

[0090] An exemplary embodiment of the present invention includes constructing a feed quality evaluation model comprising:

[0091]

[0092] In the formula, As an evaluation index, The standard particle size uniformity index. For standard mixing uniformity, This represents the current particle size uniformity index. Given the current moisture content, Standard moisture content, The total number of particles, The total number of complete particles, , , and The weights are calculated, and their sum is 1.

[0093] In this embodiment, various data are compared with standard data to obtain the deviation rate. Different weights are set for each data type to obtain a relatively objective comprehensive value.

[0094] In one exemplary embodiment of the present invention, the construction of the device control model includes:

[0095]

[0096] In the formula, For the first Equipment control index of each device. The frequency of the current device's occurrence. This represents the number of target abnormal data associated with the current device. This represents the number of normal data items currently associated with the device. The adjustment time required for the current equipment. For the first The deviation rate of the target abnormal data.

[0097] The above model considers the impact of various parameters on the final result. Combined with the impact of the current equipment adjustment on various parameters, the ranking result is obtained. The ranking result assists the system or manual adjustment. Secondly, in this embodiment, the calculation of units is not involved, only the corresponding data is taken. The obtained equipment adjustment index is only used for ranking. When calculating the equipment adjustment index of all equipment, all data can use the same unit standard.

[0098] A real-time feed quality control system based on multimodal data, used to execute the aforementioned real-time feed quality control method based on multimodal data, includes:

[0099] The data processing module is configured to acquire multimodal data of the current feed, calculate the deviation rate of each data type in the multimodal data, construct a feed quality evaluation model, output a quality index of the current feed based on the deviation rate of each data type, determine whether the current feed is normal based on the quality index; if normal, no processing is performed; if abnormal, a deviation threshold is set, and data with a deviation rate greater than the deviation threshold are designated as target abnormal data; construct a data-equipment mapping table, acquire the target equipment corresponding to each type of target abnormal data, obtain the frequency of occurrence of each target equipment, construct an equipment control model, and output an equipment control index based on the frequency and the deviation rate of the data corresponding to the current target equipment.

[0100] The execution module is configured to sort the control indices of the devices, send a control signal to the first target device in the sort, and after receiving a feedback signal that the first target device has completed the control, acquire the quality index again. If the quality index is normal, the control is stopped; if it is not normal, the control signals are sent sequentially to the next target device until the quality index is normal. If the quality index is still not normal after all target devices have been controlled, an alarm signal is issued and a control signal to stop all devices is output.

[0101] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0102] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. This computer software product, stored in a storage medium, includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0103] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for real-time control of feed quality based on multimodal data, characterized in that, include: Acquire multimodal data of the current feed, including particle size uniformity index, mixing uniformity, moisture content and particle integrity. Calculate the deviation rate of each data point in the multimodal data, construct a feed quality evaluation model, output the quality index of the current feed based on the deviation rate of each data point through the feed quality evaluation model, and determine whether the current feed is normal based on the quality index. If it is normal, no action is taken; if it is abnormal, a deviation threshold is set, and data with a deviation rate greater than the deviation threshold is taken as target abnormal data. Construct a data-device relationship mapping table, obtain the target device corresponding to each type of target abnormal data, obtain the frequency of occurrence of each target device, construct a device control model, and output the device control index based on the frequency and the deviation rate of the data corresponding to the current target device through the device control model; The control indices of the devices are sorted. A control signal is sent to the target device ranked first. After receiving a feedback signal that the target device ranked first has completed the control, the quality index is acquired again. If the quality index is normal, the control is stopped. If it is not normal, the control signal is sent sequentially to the next target device until the quality index is normal. If the quality index is still not normal after all target devices have been controlled, an alarm signal is issued and a control signal to stop all devices is output.

2. The method for real-time control of feed quality based on multimodal data according to claim 1, characterized in that, The acquisition of multimodal data of the current feed includes: Determine whether the types of multimodal data to be detected match the preset detection types. If yes, collect data normally. If no, obtain the currently missing types and collect the data corresponding to the currently missing types again. If data corresponding to the currently missing type is collected, collection proceeds normally; if data corresponding to the currently missing type is still not collected, an early warning signal is sent to the control terminal.

3. The method for real-time control of feed quality based on multimodal data according to claim 2, characterized in that... The particle size uniformity index includes: In the formula, The particle size uniformity index is... For the first Sample quality on the sieve For the first The average pore size on the sieve, The geometric mean particle size.

4. The method for real-time control of feed quality based on multimodal data according to claim 3, characterized in that, The particle integrity includes: Acquire the image data of the current feed, construct an image recognition model, identify the feed particles in the current image data using the image recognition model, and output the outline of the feed particles; Extract the target contour image of each feed pellet, compare it with the standard contour image of a standard feed pellet to obtain the similarity, set a similarity threshold, and output complete feed pellets and incomplete feed pellets based on the similarity and similarity threshold. The pellet integrity is obtained based on the number of complete and incomplete feed pellets.

5. The method for real-time control of feed quality based on multimodal data according to claim 4, characterized in that, The image recognition model is constructed by: After adaptive grayscale normalization of the image data, bilateral filtering is performed to obtain the image to be identified; The system captures the edges of particles of different sizes in the image to be identified, sets the condition for judging points on the edges as contour points, obtains several contour points, smooths and completes these contour points, and outputs the complete contour.

6. The method for real-time control of feed quality based on multimodal data according to claim 5, characterized in that, The process of obtaining the image to be identified by bilateral filtering includes: In the formula, The pixels of the image to be identified after bilateral filtering grayscale value, For pixels grayscale value at that location Spatial weights, For grayscale weights, For direction weights, For space core, For grayscale kernel, For pixels With pixels The angle between the line connecting the particles and the principal direction of the particle. For pixels The grayscale value at that location.

7. The method for real-time control of feed quality based on multimodal data according to claim 6, characterized in that, The process of capturing the edges of particles of different sizes in the image to be identified and setting the conditions for determining points on the edges as contour points includes: Perform multi-scale gradient calculation: The condition for a point to be a contour point is to satisfy the following conditions: In the formula, To fuse the magnitude of the gradient, For scale weights, For the scale is Gaussian smoothed image gradient, Let be the standard deviation of the k-th Gaussian smoothed scale. To increase the gradient threshold, The global gradient mean. The average gray level of the background. The grayscale difference threshold. The standard deviation of the background grayscale.

8. The method for real-time control of feed quality based on multimodal data according to claim 7, characterized in that, The constructed feed quality evaluation model includes: In the formula, As an evaluation index, The standard particle size uniformity index. For standard mixing uniformity, This represents the current particle size uniformity index. Given the current moisture content, Standard moisture content, The total number of particles, The total number of complete particles, , , and The weights are calculated, and their sum is 1.

9. A method for real-time control of feed quality based on multimodal data according to claim 7, characterized in that, The constructed equipment control model includes: In the formula, For the first Equipment control index for each device. The frequency of the current device's occurrence. This represents the number of target abnormal data associated with the current device. This represents the number of normal data items currently associated with the device. The adjustment time required for the current equipment. For the first The deviation rate of the target abnormal data.

10. A real-time feed quality control system based on multimodal data, characterized in that, A method for real-time feed quality control based on multimodal data as described in any one of claims 1-9 includes: The data processing module is configured to acquire multimodal data of the current feed, including particle size uniformity index, mixing uniformity, moisture content, and particle integrity; calculate the deviation rate of each data type in the multimodal data; construct a feed quality evaluation model; output the quality index of the current feed based on the deviation rate of each data type; determine whether the current feed is normal based on the quality index; if normal, no processing is performed; if abnormal, a deviation threshold is set, and data with a deviation rate greater than the deviation threshold are designated as target abnormal data; construct a data-equipment mapping table; acquire the target equipment corresponding to each type of target abnormal data; obtain the frequency of occurrence of each target equipment; construct an equipment control model; and output an equipment control index based on the frequency and the deviation rate of the data corresponding to the current target equipment. The execution module is configured to sort the control indices of the devices, send a control signal to the first target device in the sort, and after receiving a feedback signal that the first target device has completed the control, acquire the quality index again. If the quality index is normal, the control is stopped; if it is not normal, the control signals are sent sequentially to the next target device until the quality index is normal. If the quality index is still not normal after all target devices have been controlled, an alarm signal is issued and a control signal to stop all devices is output.

Images