An intelligent and automated production control system for instant rice

Abstract

The present invention provides an intelligent automated production control system for instant rice, belonging to the field of automated control technology. The system comprises: a parameter comparison and analysis module for monitoring the operating parameters of production equipment involved in each production process on the instant rice automatic production line, performing comparative analysis with corresponding standard parameters, and determining the abnormality coefficient of each operating parameter; an image monitoring module for monitoring the second status image of the production equipment before and after the execution of the matched production process; an abnormality determination module for determining the abnormal state of the corresponding production equipment based on the first status image, the second status image, and all abnormality coefficients involved in the corresponding production process; and an intelligent automated control module for determining the parameters to be adjusted of the instant rice automatic production line and the allowable deviation correction time for each parameter to be adjusted according to all abnormal states, and transmitting the information to the corresponding control components for automated control and adjustment. This effectively ensures the taste of the rice and improves the efficiency of automated production.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic control, and in particular to an intelligent automatic production control system for instant rice. Background Art

[0002] With the improvement of living standards and the acceleration of work pace, more and more people tend to eat together to save time. However, most canteens, restaurants, and hotels still use large pots to steam rice. Although each pot can produce 50-100 kg of rice, the overall rice output is slow, production efficiency is low, and the rice tastes poor and can even be undercooked. Therefore, automatic instant rice production lines have emerged.

[0003] The entire process of the instant rice automatic production line is electrically controlled, with production capacities of 150kg / h, 300kg / h, 450kg / h, and 600kg / h. However, during the rice production process, since the production line is automated and relies on fixed steaming parameters, it cannot adjust in real time based on variables such as rice quantity and moisture content. This results in large fluctuations in the quality of different batches of rice. Furthermore, if any abnormality occurs in any link, the taste of the rice produced may deteriorate, indirectly reducing the efficiency of automated production. Furthermore, the fully electrical control system lacks an intelligent pre-judgment mechanism, requiring only passive shutdown and maintenance after a fault occurs. It is unable to provide early warnings and dynamically adjust the production rhythm through anomaly detection models. Traditional control logic struggles to overcome physical production capacity limitations, and AI algorithms cannot be used to optimize the connection between various processes to achieve super-linear efficiency improvements. The lack of an intelligent energy management system prevents accurate matching of energy supply under different production modes, resulting in high unit energy consumption.

[0004] Therefore, the present invention proposes an intelligent automatic production control system for instant rice. Summary of the Invention

[0005] The present invention provides an intelligent automated production control system for instant rice, which is used to monitor the working parameters involved in each production process and determine abnormal conditions in combination with the current status diagram before and after the production process is executed. Then, by setting a reasonable allowable correction time, the adequacy of abnormal adjustments is ensured, thereby effectively ensuring the taste of the rice and improving the efficiency of automated production.

[0006] The present invention provides an intelligent automatic production control system for instant rice, comprising:

[0007] The parameter comparison and analysis module is used to monitor the working parameters of the production equipment involved in each production process of the instant rice automatic production line, and compare and analyze the working parameters with the corresponding standard parameters to determine the abnormal coefficient of each working parameter;

[0008] An image monitoring module, configured to monitor a first status image of the target rice before and after the corresponding production equipment executes the matched production process;

[0009] an abnormality determination module, configured to obtain an abnormal state of a corresponding production device based on the first status map, the second status map, and all abnormality coefficients involved in a corresponding production process;

[0010] The intelligent automation control module is used to determine the parameters to be adjusted of the instant rice automatic production line and the allowable correction time of each parameter to be adjusted according to all abnormal conditions, and send them to the corresponding control components to automatically control and adjust the corresponding production equipment.

[0011] Preferably, the production equipment includes: a rice lifting machine, a metering machine, an automatic rice washing machine, an automatic filling machine, a rice stewing machine, a rice loosening machine and a continuous pot washing machine.

[0012] Preferably, the parameter comparison and analysis module includes:

[0013] a power consumption monitoring unit, configured to monitor the power consumption of the first device at each working time point when the instant rice automated production line is in operation, wherein the first device is a device that sets the power consumption state of the production equipment to a high power consumption state;

[0014] The loss determination unit is used to determine the self-loss function ZS of each first device according to the working conditions of each first device and the power consumption at different working time points, and in combination with all working parameters involved in the first device at each working time point. i1 =Z0(working conditions, power consumption, working parameters);

[0015] a vector generating unit, configured to obtain an amplitude value of the first device based on the currently collected vibration noise data of the first device and generate a noise difference vector by combining the amplitude value with the standard noise data;

[0016] an adjusting unit, configured to adjust the noise difference vector according to a noise source corresponding to the first device to obtain a noise impact factor;

[0017]

[0018] Zs j1 =Cs j1 ×(1-G(r i1 ,f i1 ))

[0019] Among them, S i1 represents the noise impact factor of the first device i1; m i1 Zs represents the number of vibration noise data collected for the i1th first device;j1 Cs represents the value of the j1th element in the adjustment vector of the i1th first device; j1 represents the value of the j1th element in the noise difference vector of the i1th first device; G(r i1 ,f i1 ) represents the noise source r for the i1th first device i1 With noise type f i1 The error function of

[0020] Based on the self-loss function ZS i1 The relationship between the corresponding working parameters and the corresponding standard parameters is combined with the comparative analysis results and noise impact factors to determine the abnormal coefficient of the corresponding working parameters;

[0021]

[0022] Among them, Y i1,j2 Indicates the abnormal coefficient of the j2th working parameter involved in the i1th first device; s1 i1,j2 Indicates the value of the j2th operating parameter involved in the i1th first device; s0 i1,j2 represents the value of the j2th standard parameter involved in the i1th first device; represents the loss normalization coefficient of the i1th first device; τ i1,j2 Represents the self-loss function ZS i1 Conversion coefficients in relation to corresponding operating parameters;

[0023] The coefficient calculation unit is used to regard the equipment with low power consumption in the production equipment as the second equipment, and calculate the power consumption of the equipment according to the Determine the abnormal coefficient of the operating parameter involved in each second device, where Y i2,j3 q1 represents the abnormal coefficient of the j3th working parameter involved in the i2th second device; i2,j3 represents the value of the j3th operating parameter involved in the i2th second device; q0 i2,j3 Indicates the value of the j3th standard parameter involved in the i2th second device.

[0024] Preferably, the abnormality determination module includes:

[0025] a feature recognition unit configured to perform image adjustment on the second status image, match an image extraction network for the production process from a process-image database, and perform feature recognition on the first status image and the adjusted second status image involved in the corresponding production process to obtain difference features between the first and second status images;

[0026] A feature comparison and analysis unit, configured to compare and analyze the before-after difference feature with the standard difference feature of the corresponding production process to obtain a feature difference coefficient;

[0027] The state determination unit is used to determine the abnormal state of the corresponding production equipment based on the characteristic difference coefficient and in combination with all related abnormal coefficients.

[0028] Preferably, the feature recognition unit includes:

[0029] A position alignment subunit, configured to align the first current image and the second current image at the same position, and to establish a comparison array of points at the same position;

[0030] a statistical subunit, configured to count a second number of inconsistent pixel information between two pixels in the comparison array and determine a distribution state of the second number, wherein the distribution state includes: uneven distribution and relatively uniform distribution;

[0031] a radiation point determination subunit, configured to determine, from the second status map, surrounding radiation points of each pixel point in the second number, wherein the surrounding radiation points are points in the first number and are distributed around the corresponding pixel points in the second number at the closest distance;

[0032] a point determination subunit, configured to select, from all surrounding radiation points, a point having a maximum distance from a corresponding pixel point in the second number, perform circular segmentation on the second current image, and compare and analyze actual pixel information of the segmented circle with theoretical pixel information of the segmented circle at the location to determine whether the corresponding pixel point in the second number is a noise pixel point;

[0033] an information adjustment subunit, configured to, if the corresponding pixel point in the second number is a noise pixel point, extract pixel information of the corresponding pixel point from the theoretical pixel information and adjust and retain the pixel information of the corresponding pixel point in the second number in combination with the distribution state of the second number;

[0034] If not, retain the pixel information of the corresponding pixel point in the second number;

[0035] The feature recognition subunit is used to perform feature recognition on the adjusted second current image and the first current image according to the image extraction network to determine the difference features between the before and after images.

[0036] Preferably, the point judgment subunit includes:

[0037] A boundary line locking block is used to lock the theoretical boundary line in the theoretical pixel information and, at the same time, lock the actual boundary line in the actual pixel information;

[0038] a length determination block, configured to determine an overlapping boundary length Cb between the theoretical boundary line and the actual boundary line, a burr line length Zz of the corresponding pixel point in the second number, and a total burr line length Zc of the actual boundary line based on the theoretical boundary line, and determine whether the corresponding pixel point is a noise pixel point;

[0039]

[0040] Among them, Pd represents the judgment result. When the value of Pd is 1, the corresponding pixel point in the second number is determined to be a noise pixel point; when the value of Pd is 0, the corresponding pixel point in the second number is determined to be not a noise pixel point; ΔL represents the set unit length for each pixel point; r is the length value between the corresponding pixel point in the second number and the point with the maximum distance; Mb represents the total number of pixel points in the segmented circle based on the second current image; Nb represents the first number in which two pixel information of the segmented circle based on the second current image are consistent.

[0041] Preferably, the intelligent automation control module includes:

[0042] A vector construction unit is used to construct an abnormality vector based on the abnormal state of each production equipment and the abnormality coefficient of each working parameter involved;

[0043] A unit to be adjusted, used for inputting the abnormal vector into the abnormal analysis model, automatically obtaining the parameters to be adjusted and the adjustment instructions, wherein the parameters to be adjusted are the parameters to be adjusted;

[0044] The curve determination unit is used to obtain the latest maintenance plan of each production equipment and the trial operation work test data after maintenance according to the latest maintenance plan from the historical maintenance database, and obtain the trial operation working curve of each parameter to be adjusted;

[0045] A curve state analysis unit is used to analyze the curve state of the trial operation working curve, and if it is a stable state, a preset time length is used as the allowable correction time length of the corresponding parameter to be adjusted;

[0046] If it is a fluctuating state, determine the number of peaks that exist and determine the allowable correction time.

[0047] Preferably, the intelligent automation control module further includes:

[0048] a quantity classification analysis unit, configured to assign a continuous adjustment label to the corresponding parameter to be adjusted if the number of peaks is 0 after determining the number of peaks;

[0049] If the number of peaks is 1, the first duration between the start of the trial run and the peak appearance time is obtained, and a first duration period label is assigned to the corresponding parameter to be adjusted;

[0050] If the number of peaks is 2, obtain the second duration T2 between the start of the trial run and the first peak, and the third duration T3 between the two peaks. According to the result of min(T2, T3), assign the result duration period label to the corresponding parameter to be adjusted.

[0051] If the number of peaks is greater than 2, the third duration of the adjacent interval peaks is obtained and the

[0052] The result of assigning the final duration period label to the corresponding parameter to be adjusted; And A1 represents the comparison function, T3ave represents the average value of all third durations; represents the variance of all third durations; Tmin represents the minimum value among all third durations; Tmax represents the maximum value among all third durations;

[0053] The correction and adjustment unit is used to send the adjustment instructions to the corresponding production equipment to adjust the correction parameters to be adjusted according to the label assignment results, so as to realize automatic control and adjustment, wherein the label assignment result is the set allowable correction time.

[0054] Compared with the prior art, the present invention has the following advantages:

[0055] By monitoring the working parameters involved in each production process and combining the status diagrams before and after the execution of the production process to determine the abnormal state, and then setting a reasonable allowable correction time to ensure the adequacy of the abnormal adjustment, the taste of the rice is effectively guaranteed and the efficiency of automated production is improved; on the basis of the existing automated production line, combined with artificial intelligence technologies such as AI visual inspection systems and adaptive control algorithms, the entire process from raw material identification to finished product quality inspection is realized. Intelligent, focusing on the major people's livelihood needs of eating, has broad market prospects.

[0056] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0057] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0059] Figure 1This is a structural diagram of an intelligent automated production control system for instant rice in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0061] The present invention provides an intelligent automatic production control system for instant rice. Figure 1 Shown, including:

[0062] The parameter comparison and analysis module is used to monitor the working parameters of the production equipment involved in each production process of the instant rice automatic production line, and compare and analyze the working parameters with the corresponding standard parameters to determine the abnormal coefficient of each working parameter;

[0063] An image monitoring module, configured to monitor a first status image of the target rice before and after the corresponding production equipment executes the matched production process;

[0064] an abnormality determination module, configured to obtain an abnormal state of a corresponding production device based on the first status map, the second status map, and all abnormality coefficients involved in a corresponding production process;

[0065] The intelligent automation control module is used to determine the parameters to be adjusted of the instant rice automatic production line and the allowable correction time of each parameter to be adjusted according to all abnormal conditions, and send them to the corresponding control components to automatically control and adjust the corresponding production equipment.

[0066] Preferably, the production equipment includes: a rice lifting machine, a metering machine, an automatic rice washing machine, an automatic filling machine, a rice stewing machine, a rice loosening machine and a continuous pot washing machine.

[0067] In this embodiment, the instant rice automatic production line includes:

[0068] Rice Hoist: Rice is unpacked and fed into the rice hopper. A small rice hoist connected to a belt moves upward along the belt, feeding the rice into the rice hopper chute. This rice hoist has a dual function: during storage, the rice hopper feeds the rice into the chute, and when the chute is full, the rice enters the rice hopper. During cooking, the rice meter operates, and the horizontal flap at the bottom of the rice hopper opens. The rice in the hopper is then lifted by the rice hoist into the chute for metering. The rice hopper consists of two parts: the upper part is a rectangular structure with an open chute, a top observation hole cover, and peepholes on the side. The lower part is a funnel-shaped structure connected to the bucket rice hoist, and the entire structure is a steel frame.

[0069] Metering Machine: This is a volumetric metering machine consisting of a metering rotor and a machine body. The metering amount is adjusted using an adjustment lever on the rotor. Each rotation of the rotor completes the rice drop process, and after seven rotations, the metering machine produces exactly 7.5 kg of rice per pot. The PLC and a cam travel switch on the rotor control each metering machine operation, sending a signal to the human-machine interface to record the number of pots cooked during that shift.

[0070] Automatic rice washing machine: The rice washing machine consists of a frame and a rice washing trough, a rice pumping system, and a water distribution system. The rice is fully cleaned by a reciprocating conveying and flushing method. The cleaned rice is then pumped into a filling tank for the next process.

[0071] Automatic filling machine: It consists of a machine frame, soaking tank, metering mechanism, rice feeding system, and water adding system. After the rice is added to the rice cooker, it is pushed to the water dispensing position for water dispensing. After the water is dispensed, the rice cooker is sent to the soaking track and automatically enters the rice cooking chamber.

[0072] The rice cooker consists of a rice cooking and transport system, a combustion system, a pot hoist, a rice cooking transport, and a frame. The rice cooking and transport drive features stepless speed adjustment based on rice quality and heat level, typically cooking for approximately 17 minutes. The combustion system is equipped with 10 high-density, slanted burning plates and corresponding air inlet control valves, allowing the size and intensity of the flames on each plate to be adjusted according to cooking requirements. After cooking is complete, a hoisting mechanism raises the pot to the upper level of the cooking chamber, where it is cooked using residual heat. The rice cooker consists of a frame and a rice cooking and transport track. The pot slowly moves along this track, cooking for approximately 17 minutes, ensuring the rice is fully gelatinized and resulting in a rich, fragrant, and delicious rice.

[0073] Rice loosening machine: This machine consists of several parts including turning, loosening, belt transport, and frame. The whole pot of cooked rice is pushed into the turning arm fork through an unpowered roller conveyor, turned about 135 degrees and poured into the loosening trough. The rice balls are evenly loosened into a loose shape by the staggered rotating loosening teeth, and then evenly transported out through the food conveyor belt.

[0074] Continuous pot washing machine: The pot washing machine consists of a pot transport system, a water tank, upper and lower hot water spray pipes, a reduction motor, a hot water pump, and a frame. Before washing, the rice pot is manually flipped and placed on the pot washing machine's entrance conveyor chain. The pot automatically enters the pot washing machine, where it is rinsed with 65°C to 70°C hot water sprayed through the upper and lower high-pressure spray pipes of the two tanks. After being rinsed, it is automatically conveyed out of the pot washing machine. After washing, it is manually removed from the end of the conveyor chain and conveyed to the rice washing machine, where it is then recycled on the unpowered rice receiving and water distribution rollers.

[0075] In this embodiment, the working parameters refer to the various parameters involved in the operation of the corresponding equipment, such as the transmission speed of the belt, the opening size of the horizontal lower plug plate, the adjustment metering of the metering rotor, the injection volume of the pump, the amount of automatic water distribution, etc. The standard parameters are pre-set, that is, they are determined before the production line is put into use to ensure that the production line can operate effectively.

[0076] In this embodiment, for example, before and after the execution of the process corresponding to the rice lifting machine, the first status image is an image taken of the rice bin chute before the rice is sent into the rice bin chute as the belt moves upward, and the second status image is an image taken of the rice bin chute after the rice is sent into the rice bin chute as the belt moves upward. And so on, each production process will have corresponding pre-execution images and post-execution images.

[0077] In this embodiment, the abnormal state is obtained based on the characteristic difference coefficient and combined with the abnormal coefficient analysis.

[0078] In this embodiment, the parameters to be adjusted refer to the parameters that need to be adjusted. However, since some parameters need to be adjusted all the time, some parameters may not need to be adjusted after the current adjustment. Therefore, it is necessary to set the allowed correction time to make corresponding adjustments to different parameters to ensure the reliable operation of the production equipment to the greatest extent and realize automatic control and adjustment. Among them, the parameters to be adjusted can be the running speed of the belt, the cooking time, the strength of the burning plate flame, etc.

[0079] The beneficial effect of the above technical solution is: by monitoring the working parameters involved in each production process, and combining the current status diagram before and after the execution of the production process to determine the abnormal state, and then by setting a reasonable allowable correction time to ensure the adequacy of the abnormal adjustment, effectively ensure the taste of the rice, and improve the efficiency of automated production.

[0080] The present invention provides an intelligent automatic production control system for instant rice, wherein the parameter comparison and analysis module comprises:

[0081] a power consumption monitoring unit, configured to monitor the power consumption of the first device at each working time point when the instant rice automated production line is in operation, wherein the first device is a device that sets the power consumption state of the production equipment to a high power consumption state;

[0082] The loss determination unit is used to determine the self-loss function ZS of each first device according to the working conditions of each first device and the power consumption at different working time points, and in combination with all working parameters involved in the first device at each working time point. i1 =Z0(working conditions, power consumption, working parameters);

[0083] a vector generating unit, configured to obtain an amplitude value of the first device based on the currently collected vibration noise data of the first device and generate a noise difference vector by combining the amplitude value with the standard noise data;

[0084] an adjusting unit, configured to adjust the noise difference vector according to a noise source corresponding to the first device to obtain a noise impact factor;

[0085]

[0086] Zs j1 =Cs j1 ×(1-G(r i1 ,f i1 ))

[0087] Among them, S i1 represents the noise impact factor of the first device i1; m i1 Zs represents the number of vibration noise data collected for the i1th first device; j1 Cs represents the value of the j1th element in the adjustment vector of the i1th first device; j1 represents the value of the j1th element in the noise difference vector of the i1th first device; G(r i1 ,f i1 ) represents the noise source r for the i1th first device i1 With noise type f i1 The error function of

[0088] Based on the self-loss function ZS i1 The relationship between the corresponding working parameters and the corresponding standard parameters is combined with the comparative analysis results and noise impact factors to determine the abnormal coefficient of the corresponding working parameters;

[0089]

[0090] Among them, Y i1,j2 Indicates the abnormal coefficient of the j2th working parameter involved in the i1th first device; s1 i1,j2 Indicates the value of the j2th operating parameter involved in the i1th first device; s0 i1,j2 represents the value of the j2th standard parameter involved in the i1th first device; represents the loss normalization coefficient of the i1th first device; τ i1,j2 Represents the self-loss function ZS i1 Conversion coefficients in relation to corresponding operating parameters;

[0091] The coefficient calculation unit is used to regard the equipment with low power consumption in the production equipment as the second equipment, and calculate the power consumption of the equipment according to the Determine the abnormal coefficient of the operating parameter involved in each second device, where Y i2,j3 q1 represents the abnormal coefficient of the j3th working parameter involved in the i2th second device; i2,j3 represents the value of the j3th operating parameter involved in the i2th second device; q0 i2,j3 Indicates the value of the j3th standard parameter involved in the i2th second device.

[0092] In this embodiment, the operating state refers to the state after the production line is put into operation and is in a stable state. Therefore, under normal circumstances, the production line will be unstable at the beginning of operation, and there will be a buffer transition stage. For example, after running for 3 minutes, it is considered to enter a stable state. At this time, the power consumption is monitored.

[0093] In this embodiment, the high power consumption state refers to the power consumption being greater than the set power threshold. At this time, for example, the rice cooker is a device in the high power consumption state and is regarded as the first device. The low power consumption state refers to the device whose power consumption is not greater than the set power threshold and is regarded as the second device.

[0094] In this embodiment, the working conditions are set before the production line is put into use and are related to the standard power consumption of the standard operating conditions of different components. For example, the standard speed is a1, the power consumption is R1, the flame strength of the burning plate is b1, the power consumption is R2, etc., which is mainly used as a reference for subsequent loss analysis.

[0095] In this embodiment, the self-loss function ZS of the first device based on working conditions, power consumption, and working parameters is matched from the device-condition-power consumption-parameter-loss function database. i1 It should be noted that the database contains functions of combined power consumption and combined working parameters under different working conditions, which can be obtained directly. Generally speaking, the greater the power consumption, the greater the difference between the working parameters and the standard parameters under the set working conditions, and the greater the corresponding self-loss.

[0096] In this embodiment, the vibration noise data refers to the vibration amplitude of the device, because the device may have a certain vibration or no vibration during operation. When there is no vibration, the amplitude value is replaced by 0. When there is vibration, the amplitude value is represented by the actual value.

[0097] In this embodiment, the standard noise data is obtained before the production line is put into use, and the noise difference vector = {each amplitude value corresponding to the first device - the value of the standard noise data}.

[0098] In this embodiment, G(r i1 ,f i1 ) ranges from 0 to 0.1 and is obtained from the error-value comparison table, and the noise source ri1 With noise type f i1 It is the error value directly obtained by comparative analysis from the table, which is pre-set.

[0099] In this embodiment, the mutual relationship refers to whether the working component affected by the working parameters will cause power loss. If so, there is a mutual relationship; otherwise, there is no mutual relationship, and it is obtained by matching the parameter-relationship comparison table, which contains the working parameters of different components and the corresponding power loss, and is set in advance.

[0100] The beneficial effects of the above technical solution are: measuring the self-loss function from three aspects: power consumption, working conditions and working parameters, and combining the noise impact factors and mutual relationships determined by the noise difference vector to comprehensively determine the abnormal coefficient of the equipment in the high power consumption state, and determining the abnormal coefficient of the equipment in the low power consumption state by the difference between the actual and standard parameters, to ensure the rationality of the coefficient set for each device, and provide a basis for subsequent analysis.

[0101] The present invention provides an intelligent automatic production control system for instant rice, wherein the abnormality determination module comprises:

[0102] a feature recognition unit configured to perform image adjustment on the second status image, match an image extraction network for the production process from a process-image database, and perform feature recognition on the first status image and the adjusted second status image involved in the corresponding production process to obtain difference features between the first and second status images;

[0103] A feature comparison and analysis unit, configured to compare and analyze the before-after difference feature with the standard difference feature of the corresponding production process to obtain a feature difference coefficient;

[0104] The state determination unit is used to determine the abnormal state of the corresponding production equipment based on the characteristic difference coefficient and in combination with all related abnormal coefficients.

[0105] In this embodiment, the purpose of image adjustment is to ensure that the image is more consistent with the current actual situation after the process.

[0106] In this embodiment, the process-image database includes different production processes and networks for extracting image features for the processes. This is because different network applications use different image scenes, which means that the features extracted from the images need to fit the corresponding process itself, so as to determine whether the process operation is reasonable.

[0107] In this embodiment, the before-after difference feature refers to the different results after the process is executed.

[0108] In this embodiment, the feature difference coefficient=(feature difference between the before-after difference feature and the standard difference feature) / standard difference feature.

[0109] In this embodiment, the standard difference feature is preset, that is, the feature determined by the difference between the images obtained before and after the process is executed is used as a standard for determining a coefficient.

[0110] In this embodiment, the abnormal state is obtained by matching the characteristic difference coefficient and the abnormal coefficient involved in the production equipment of the production process from the equipment-coefficient combination-state comparison table. The comparison table contains the characteristic difference coefficients, abnormal coefficients and abnormal states matching different combinations. Among them, the abnormal state can be the operation abnormality of a certain component, such as the abnormality of the automatic rice washing machine not cleaning cleanly, the abnormality of insufficient water for cleaning, the abnormal spray force of the pump, etc.

[0111] The beneficial effect of the above technical solution is: by adjusting the second status map and combining the network to perform feature recognition, the before and after difference features are obtained, and then compared with the standard difference features to obtain the feature difference coefficient, and combined with the abnormal coefficient, the abnormal state is comprehensively determined to provide a basis for subsequent analysis.

[0112] The present invention provides an intelligent automatic production control system for instant rice, wherein the feature recognition unit comprises:

[0113] A position alignment subunit, configured to align the first current image and the second current image at the same position, and to establish a comparison array of points at the same position;

[0114] a statistical subunit, configured to count a second number of inconsistent pixel information between two pixels in the comparison array and determine a distribution state of the second number, wherein the distribution state includes: uneven distribution and relatively uniform distribution;

[0115] a radiation point determination subunit, configured to determine, from the second status map, surrounding radiation points of each pixel point in the second number, wherein the surrounding radiation points are points in the first number and are distributed around the corresponding pixel points in the second number at the closest distance;

[0116] a point determination subunit, configured to select, from all surrounding radiation points, a point having a maximum distance from a corresponding pixel point in the second number, perform circular segmentation on the second current image, and compare and analyze actual pixel information of the segmented circle with theoretical pixel information of the segmented circle at the location to determine whether the corresponding pixel point in the second number is a noise pixel point;

[0117] an information adjustment subunit, configured to, if the corresponding pixel point in the second number is a noise pixel point, extract pixel information of the corresponding pixel point from the theoretical pixel information and adjust and retain the pixel information of the corresponding pixel point in the second number in combination with the distribution state of the second number;

[0118] If not, retain the pixel information of the corresponding pixel point in the second number;

[0119] The feature recognition subunit is used to perform feature recognition on the adjusted second current image and the first current image according to the image extraction network to determine the difference features between the before and after images.

[0120] Preferably, the point judgment subunit includes:

[0121] A boundary line locking block is used to lock the theoretical boundary line in the theoretical pixel information and, at the same time, lock the actual boundary line in the actual pixel information;

[0122] a length determination block, configured to determine an overlapping boundary length Cb between the theoretical boundary line and the actual boundary line, a burr line length Zz of the corresponding pixel point in the second number, and a total burr line length Zc of the actual boundary line based on the theoretical boundary line, and determine whether the corresponding pixel point is a noise pixel point;

[0123]

[0124] Among them, Pd represents the judgment result. When the value of Pd is 1, the corresponding pixel point in the second number is determined to be a noise pixel point; when the value of Pd is 0, the corresponding pixel point in the second number is determined to be not a noise pixel point; ΔL represents the set unit length for each pixel point; r is the length value between the corresponding pixel point in the second number and the point with the maximum distance; Mb represents the total number of pixel points in the segmented circle based on the second current image; Nb represents the first number in which two pixel information of the segmented circle based on the second current image are consistent.

[0125] In this embodiment, the theoretical boundary line is determined based on theoretical pixel information, and the actual boundary line is determined based on current pixel information, which is implemented based on an image edge retrieval algorithm.

[0126] In this embodiment, the burr line refers to a bifurcation line corresponding to the actual boundary line, and the bifurcation line is a thin line.

[0127] In this embodiment, the comparison array = {pixel information of the same position point based on the first current map and pixel information of the same position point based on the second current map}.

[0128] In this embodiment, the distribution state refers to the position distribution of the position points of inconsistent pixel information based on the second status map, wherein relatively uniform distribution means that the distribution density of the position points is within the set density difference variation range, which can be regarded as relatively uniform.

[0129] In this embodiment, the surrounding radiation points refer to points in different surrounding directions surrounding each pixel point in the second number and presenting the points with the closest distance distribution.

[0130] In this embodiment, the theoretical pixel information refers to the standard pixel information theoretically obtained before and after the execution of the corresponding process. For example, for an image of rice washing, the theoretical image after washing is pre-set.

[0131] In this embodiment, if it is a noise pixel:

[0132] If the distribution state of the second number is relatively uniform, then the pixel information of the corresponding pixel point in the second number is replaced based on the pixel information of the pixel point extracted based on the theoretical pixel information;

[0133] If the distribution state of the second number is uneven, then the pixel information of the pixel points extracted based on the theoretical pixel information is replaced with the average of the pixel information of the corresponding pixel points in the second number.

[0134] The beneficial effect of the above technical solution is: the current status map is aligned with the same position points to determine the distribution state of the second number, and the surrounding radiation points of the pixel point are combined to determine whether it is a noise pixel point, thereby facilitating the adjustment of pixel information and avoiding the reduction of information accuracy due to the presence of noise, to ensure the reliability of the difference feature determination, among which, by performing comparative analysis of the boundary line to calculate and determine whether it is a noise pixel point, providing a basis for subsequent information adjustment.

[0135] The present invention provides an intelligent automatic production control system for instant rice, wherein the intelligent automatic control module comprises:

[0136] A vector construction unit is used to construct an abnormality vector based on the abnormal state of each production equipment and the abnormality coefficient of each working parameter involved;

[0137] A unit to be adjusted, used for inputting the abnormal vector into the abnormal analysis model, automatically obtaining the parameters to be adjusted and the adjustment instructions, wherein the parameters to be adjusted are the parameters to be adjusted;

[0138] The curve determination unit is used to obtain the latest maintenance plan of each production equipment and the trial operation work test data after maintenance according to the latest maintenance plan from the historical maintenance database, and obtain the trial operation working curve of each parameter to be adjusted;

[0139] A curve state analysis unit is used to analyze the curve state of the trial operation working curve, and if it is a stable state, a preset time length is used as the allowable correction time length of the corresponding parameter to be adjusted;

[0140] If it is a fluctuating state, determine the number of peaks that exist and determine the allowable correction time.

[0141] In this embodiment, the abnormality vector = {the abnormal state of each production equipment and the abnormality coefficient of the related working parameters}.

[0142] In this embodiment, the abnormality analysis model is obtained by training a neural network model based on samples of different abnormality vectors and analysis results of the vector abnormalities by maintenance experts (parameters that need to be adjusted and adjustment instructions).

[0143] In this embodiment, the historical maintenance database contains the maintenance status of different production equipment on the production line. Since the equipment will have a certain trial phase after maintenance, the trial operation test data can be directly obtained, that is, the value of different parameters in the trial operation phase.

[0144] In this embodiment, the preset duration may be 1 minute.

[0145] In this embodiment, for example, the transmission speed of the conveyor belt needs to be adjusted. At this time, it is necessary to send instructions to the motor that controls the conveyor belt transportation, and control the transmission speed by controlling the rotation speed of the motor. At this time, after obtaining the trial operation test data for the transmission speed of the conveyor belt, a speed curve is constructed. If the speed remains unchanged, if there is a speed abnormality later, the motor is controlled to be adjusted according to the preset time length to ensure that the motor can be adjusted within the time length to ensure a certain transmission speed of the conveyor belt.

[0146] In this embodiment, the number of peaks refers to the number determined by plotting a curve for each parameter based on the test data collected during the trial operation.

[0147] The beneficial effects of the above technical solution are: by constructing an abnormality vector and combining it with the analysis model to obtain the parameters to be adjusted and the adjustment instructions, and in order to ensure that the parameter adjustment time is sufficient to ensure the reliable operation of the equipment, the allowable correction time is set for different parameters to be adjusted, one is to ensure the reliability of the control, and the other is to ensure the stable operation of the equipment.

[0148] The present invention provides an intelligent automatic production control system for instant rice, wherein the intelligent automatic control module further comprises:

[0149] a quantity classification analysis unit, configured to assign a continuous adjustment label to the corresponding parameter to be adjusted if the number of peaks is 0 after determining the number of peaks;

[0150] If the number of peaks is 1, the first duration between the start of the trial run and the peak appearance time is obtained, and a first duration period label is assigned to the corresponding parameter to be adjusted;

[0151] If the number of peaks is 2, obtain the second duration T2 between the start of the trial run and the first peak, and the third duration T3 between the two peaks. According to the result of min(T2, T3), assign the result duration period label to the corresponding parameter to be adjusted.

[0152] If the number of peaks is greater than 2, the third duration of the adjacent interval peaks is obtained and the

[0153] The result of assigning the final duration period label to the corresponding parameter to be adjusted; And A1 represents the comparison function, T3ave represents the average value of all third durations; represents the variance of all third durations; Tmin represents the minimum value among all third durations; Tmax represents the maximum value among all third durations;

[0154] The correction and adjustment unit is used to send the adjustment instructions to the corresponding production equipment to adjust the correction parameters to be adjusted according to the label assignment results, so as to realize automatic control and adjustment, wherein the label assignment result is the set allowable correction time.

[0155] In this embodiment, the continuous adjustment tag means that control instructions need to be continuously issued to the control component of the parameter to be adjusted to avoid deviation.

[0156] In this embodiment, the first duration period tag refers to the need to send control to the control component of the parameter with the first duration as a period.

[0157] In this embodiment, the result duration period tag refers to the need to send control to the control component of the parameter based on the result duration as a period.

[0158] In this embodiment, the final duration period label result duration period label refers to the control component of the parameter that needs to be issued with the final duration as a period for control.

[0159] The beneficial effects of the above technical solution are: different parameters to be adjusted are labeled and discussed in a classified manner according to the number of peak waves, which effectively ensures the reasonable control of components corresponding to different parameters to be adjusted, maximizes the normal operation of the production line, and improves automation efficiency.

[0160] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An intelligent automatic production control system for instant rice, characterized in that: include: The parameter comparison and analysis module is used to monitor the working parameters of the production equipment involved in each production process of the instant rice automatic production line, and compare and analyze the working parameters with the corresponding standard parameters to determine the abnormal coefficient of each working parameter; An image monitoring module, configured to monitor a first status image of the target rice before and after the corresponding production equipment executes the matched production process; an abnormality determination module, configured to obtain an abnormal state of a corresponding production device based on the first status map, the second status map, and all abnormality coefficients involved in a corresponding production process; An intelligent automation control module is used to determine the parameters to be adjusted of the instant rice automatic production line and the allowable correction time for each parameter to be adjusted according to all abnormal conditions, and send them to the corresponding control components to automatically control and adjust the corresponding production equipment; Wherein, the parameter comparison and analysis module includes: a power consumption monitoring unit, configured to monitor the power consumption of the first device at each working time point when the instant rice automated production line is in operation, wherein the first device is a device that sets the power consumption state of the production equipment to a high power consumption state; The loss determination unit is used to determine the self-loss function of the first device according to the working conditions of each first device and the power consumption at different working time points, and in combination with all working parameters involved in the first device at each working time point. ; a vector generating unit, configured to obtain an amplitude value of the first device based on the currently collected vibration noise data of the first device and generate a noise difference vector by combining the amplitude value with the standard noise data; an adjusting unit, configured to adjust the noise difference vector according to a noise source corresponding to the first device to obtain a noise impact factor; ; ; in, represents the noise impact factor of the i1th first device; represents the number of vibration noise data collected for the i1th first device; represents the value of the j1th element in the adjustment vector of the i1th first device; represents the value of the j1th element in the noise difference vector of the i1th first device; Indicates the noise source existing for the i1th first device and noise type The error function of Based on the self-loss function The relationship between the corresponding working parameters and the corresponding standard parameters is combined with the comparative analysis results and noise impact factors to determine the abnormal coefficient of the corresponding working parameters; ; in, represents the abnormal coefficient of the j2th operating parameter involved in the i1th first device; represents the value of the j2th operating parameter involved in the i1th first device; represents the value of the j2th standard parameter involved in the i1th first device; represents the loss normalization coefficient of the i1th first device; represents the self-loss function Conversion coefficients in relation to corresponding operating parameters; The coefficient calculation unit is used to regard the equipment with low power consumption in the production equipment as the second equipment, and calculate the power consumption of the equipment according to the Determine the abnormal coefficient of the operating parameter related to each second device, wherein, represents the abnormal coefficient of the j3th operating parameter involved in the i2th second device; represents the value of the j3th operating parameter involved in the i2th second device; represents the value of the j3th standard parameter involved in the i2th second device; The high power consumption state refers to a state where the power consumption is greater than a set power threshold, and the low power consumption state refers to a state where the power consumption is not greater than the set power threshold.

2. The intelligent automatic production control system for instant rice according to claim 1, characterized in that: The production equipment includes: a rice lifting machine, a metering machine, an automatic rice washing machine, an automatic filling machine, a rice stewing machine, a rice loosening machine and a continuous pot washing machine.

3. The intelligent automatic production control system for instant rice according to claim 1, characterized in that: The abnormality determination module includes: a feature recognition unit configured to perform image adjustment on the second status image, match an image extraction network for the production process from a process-image database, and perform feature recognition on the first status image and the adjusted second status image involved in the corresponding production process to obtain difference features between the first and second status images; A feature comparison and analysis unit, configured to compare and analyze the before-after difference feature with the standard difference feature of the corresponding production process to obtain a feature difference coefficient; The state determination unit is used to determine the abnormal state of the corresponding production equipment based on the characteristic difference coefficient and in combination with all related abnormal coefficients.

4. The intelligent automatic production control system for instant rice according to claim 3, characterized in that: The feature recognition unit includes: A position alignment subunit, configured to align the first current image and the second current image at the same position, and to establish a comparison array of points at the same position; a statistical subunit, configured to count a second number of inconsistent pixel information between two pixels in the comparison array and determine a distribution state of the second number, wherein the distribution state includes: uneven distribution and relatively uniform distribution; a radiation point determination subunit, configured to determine, from the second status map, surrounding radiation points of each pixel point in the second number, wherein the surrounding radiation points are points in the first number and are distributed around the corresponding pixel points in the second number at the closest distance; a point determination subunit, configured to select, from all surrounding radiation points, a point having a maximum distance from a corresponding pixel point in the second number, perform circular segmentation on the second current image, and compare and analyze actual pixel information of the segmented circle with theoretical pixel information of the segmented circle at the location to determine whether the corresponding pixel point in the second number is a noise pixel point; an information adjustment subunit, configured to, if the corresponding pixel point in the second number is a noise pixel point, extract pixel information of the corresponding pixel point from the theoretical pixel information and adjust and retain the pixel information of the corresponding pixel point in the second number in combination with the distribution state of the second number; If not, retain the pixel information of the corresponding pixel point in the second number; The feature recognition subunit is used to perform feature recognition on the adjusted second current image and the first current image according to the image extraction network to determine the difference features between the before and after images.

5. The intelligent automatic production control system for instant rice according to claim 4, characterized in that: Point judgment subunit, including: A boundary line locking block is used to lock the theoretical boundary line in the theoretical pixel information and, at the same time, lock the actual boundary line in the actual pixel information; a length determination block, configured to determine an overlapping boundary length Cb between the theoretical boundary line and the actual boundary line, a burr line length Zz of the corresponding pixel point in the second number, and a total burr line length Zc of the actual boundary line based on the theoretical boundary line, and determine whether the corresponding pixel point is a noise pixel point; ; Wherein, Pd represents a judgment result. When the value of Pd is 1, it is determined that the corresponding pixel point in the second number is a noise pixel point; when the value of Pd is 0, it is determined that the corresponding pixel point in the second number is not a noise pixel point; L represents the set unit length for each pixel point; r is the length value between the corresponding pixel point and the point with the maximum distance in the second number; Mb represents the total number of pixel points in the segmented circle based on the second current image; Nb represents the first number of two pixel information that are consistent in the segmented circle based on the second current image.

6. The intelligent automatic production control system for instant rice according to claim 1, characterized in that: The intelligent automation control module includes: A vector construction unit is used to construct an abnormality vector based on the abnormal state of each production equipment and the abnormality coefficient of each working parameter involved; A unit to be adjusted, used for inputting the abnormal vector into the abnormal analysis model, automatically obtaining the parameters to be adjusted and the adjustment instructions, wherein the parameters to be adjusted are the parameters to be adjusted; The curve determination unit is used to obtain the latest maintenance plan of each production equipment and the trial operation work test data after maintenance according to the latest maintenance plan from the historical maintenance database, and obtain the trial operation working curve of each parameter to be adjusted; A curve state analysis unit is used to analyze the curve state of the trial operation working curve, and if it is a stable state, a preset time length is used as the allowable correction time length of the corresponding parameter to be adjusted; If it is a fluctuating state, determine the number of peaks that exist and determine the allowable correction time.

7. The intelligent automatic production control system for instant rice according to claim 6, characterized in that: The intelligent automation control module further includes: a quantity classification analysis unit, configured to assign a continuous adjustment label to the corresponding parameter to be adjusted if the number of peaks is 0 after determining the number of peaks; If the number of peaks is 1, the first duration between the start of the trial run and the peak appearance time is obtained, and a first duration period label is assigned to the corresponding parameter to be adjusted; If the number of peaks is 2, obtain the second duration T2 between the start of the trial run and the first peak, and the third duration T3 between the two peaks. According to the result of min(T2, T3), assign the result duration period label to the corresponding parameter to be adjusted. If the number of peaks is greater than 2, the third duration of the adjacent interval peaks is obtained and the The result of assigning the final duration period label to the corresponding parameter to be adjusted; , and A1 represents the comparison function, T3ave represents the average value of all third durations; represents the variance of all third durations; Tmin represents the minimum value among all third durations; Tmax represents the maximum value among all third durations; The correction and adjustment unit is used to send the adjustment instructions to the corresponding production equipment to adjust the correction parameters to be adjusted according to the label assignment results, so as to realize automatic control and adjustment, wherein the label assignment result is the set allowable correction time.

Images