An automatic control system for soybean milk separation

Abstract

The application provides an automatic control system for soybean milk separation, which is applied to a bean product processing device and belongs to the technical field of bean product processing. A main controller is a core processing unit of the automatic control system and is internally provided with preconfigured multi-stage control logic for soybean milk separation. A sensor group is electrically connected with an input port of the main controller and is used for collecting environmental parameters in a soybean milk separation process. A man-machine interactive interface is in communication connection with the main controller and is used for receiving process parameters set by a user and displaying a working state of the automatic control system in real time. The main controller is used for analyzing the environmental parameters to obtain an initial control instruction set, adjusting the initial control instruction set according to the process parameters and the working state, and delivering the initial control instruction set to an actuator group to execute corresponding actions. The automatic control system solves the pain points of fragmented parameter control, frequent manual intervention and long production changeover debugging in traditional soybean milk separation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bean product processing, and in particular to an automatic control system for soybean milk separation. BACKGROUND

[0002] As a key link determining product quality, the technical level of soybean milk separation directly affects the protein extraction rate, product stability and production efficiency. However, the technology still has significant limitations in automatic control and parameter coordination. For example, the vibrating sieve separation equipment is the mainstream choice for small and medium-sized bean product enterprises, which realizes slurry residue separation through three-dimensional vibration of fixed frequency. However, the core parameters such as vibration frequency and amplitude need to be manually preset and fixed throughout the process. In actual production, it is found that this fixed parameter mode cannot adapt to the differences in characteristics of raw materials of different beans and different soaking degrees. When processing soybean varieties with high protein content, the sieve screen is often blocked due to insufficient vibration intensity, and excessive vibration is easy to cause protein denaturation for low water content raw materials. In addition, the temperature control of such equipment mostly relies on independent heating devices, and lacks linkage with the screening process. When the soybean milk temperature deviates from the optimal separation interval, the separation efficiency is easy to decrease or the water content of soybean residue is easy to exceed the standard.

[0003] The core defects of the prior art are mainly in three aspects: first, the parameter control is fragmented, and key parameters such as temperature, liquid level and vibration intensity are mostly controlled independently, lacking a coordinated adjustment mechanism. For example, the amplitude adjustment of the vibrating screen is irrelevant to the concentration change. Second, the self-adaptive ability is insufficient, and the process parameters cannot be automatically corrected according to the differences in bean varieties and batch changes, resulting in the need to stop production and re-adjust parameters during production change. Third, the automation degree is limited, and most equipment still relies on manual judgment of separation effect and parameter correction, which not only has high labor intensity, but also causes large fluctuations in product qualification rate due to human errors. With the improvement of consumers' requirements for the consistency of soybean milk quality and the pursuit of efficiency in industrial production, the existing technology has been difficult to meet the production needs of high precision, high stability and low manual intervention.

[0004] Therefore, the present application provides an automatic control system for soybean milk separation. SUMMARY

[0005] The present application provides an automatic control system for soybean milk separation, which is applied to a bean product processing device and comprises:

[0006] The main controller is the core processing unit of the automatic control system, and is internally provided with a preconfigured multi-stage control logic for soybean milk separation. The multi-stage control logic includes a concentration self-adaptive control stage and a temperature-liquid level coordinated control stage.

[0007] A sensor group is electrically connected to an input port of the main controller, and is configured to collect environmental parameters of the soybean milk separation process. The sensor group at least includes: a liquid level sensor configured to detect a liquid level of the soybean milk in the separation tank; a temperature sensor configured to detect a real-time temperature of the soybean milk; a viscosity detection unit configured to indirectly infer a concentration state of the soybean milk; and a photoelectric turbidity sensor configured to collect a turbidity of the soybean milk before separation, wherein the viscosity detection unit and the photoelectric turbidity sensor are configured to collect synchronously.

[0008] A human-computer interaction interface is communicatively connected to the main controller, and is configured to receive process parameters set by a user and display a working state of the automatic control system in real time.

[0009] The main controller is configured to analyze the environmental parameters to obtain an initial control instruction set. The initial control instruction set includes: an initial motor control instruction for the vibration motor driver, an initial heating control instruction for the heating control circuit, and an initial trigger opening instruction or an initial closing instruction for the water inlet electromagnetic valve for the vibration motor driver.

[0010] The initial control instruction set is adjusted according to the process parameters and the working state, and is sent to the actuator group to perform corresponding actions. The actuator group at least includes:

[0011] The vibration motor driver is configured to drive and control a vibration frequency and a vibration amplitude of the separation screen.

[0012] The heating control circuit is configured to adjust a power of the heating pipe according to a temperature feedback.

[0013] The water inlet electromagnetic valve is configured to control on-off of water supply to the separation tank.

[0014] Preferably, the main controller is configured to receive a current signal collected by the viscosity detection unit as a first feedback value representing a concentration of the soybean milk, and receive a current signal collected by the photoelectric turbidity sensor as a second feedback value representing a turbidity of the soybean milk, and determine an initial vibration frequency of the vibration motor driver.

[0015] , and ;

[0016] wherein, is the initial vibration frequency of the vibration motor driver; is a reference vibration frequency of the vibration motor driver; is the first feedback value; is a reference viscosity current value corresponding to a standard concentration; is the second feedback value; is a reference turbidity current value corresponding to a standard turbidity; and is a viscosity influence weight coefficient. is a turbidity influence weight coefficient; is a separation stage correction coefficient; is a bean species correction coefficient; is a frequency analysis function;

[0017] The main controller is configured to perform a first comparison between the first feedback values obtained at different time instants and a preset concentration-vibration parameter mapping table and a second comparison between the second feedback values obtained at different time instants and a preset turbidity-vibration parameter mapping table, and construct a two-dimensional array.

[0018] The main controller is configured to perform fitting analysis on two rows of arrays in the two-dimensional array respectively, and perform ladder analysis on the concentration fitting function and the turbidity fitting function to obtain a set of offsets.

[0019] The main controller is configured to expand the range of the initial vibration frequency in dependence on the set of offsets, and screen a middle value in the expanded range as a target vibration frequency.

[0020] The main controller is configured to obtain a target amplitude of the vibration motor driver based on the target vibration frequency

[0021] ;

[0022] wherein, is an initial amplitude of the vibration motor driver; is a reference amplitude of the vibration motor driver; is a turbidity influence coefficient on amplitude; is a frequency correction term; is a maximum vibration frequency in the two-dimensional array; is a screen aperture; is a vibration motor wheelbase length;

[0023] The main controller is configured to obtain an initial motor control instruction for controlling the vibration motor driver to work based on the target vibration frequency and the target amplitude.

[0024] Preferably, the main controller is configured to calculate an output power duty cycle of the heating control circuit.

[0025] ;

[0026] wherein, is a real-time output power duty cycle of the heating control circuit; is a proportional coefficient; is a temperature deviation; is a temperature segmentation coefficient, and ; ​is an integral time constant; is an integral correction coefficient, and ; is a differential time constant; is a feed-forward compensation coefficient; is a target temperature at a current time in a preset temperature curve; is a real-time temperature of the soy milk detected by the temperature sensor; is a real-time density of the soy milk at the current temperature; is a real-time specific heat capacity of the soy milk at the current temperature; is a standard density of the soy milk at 25°C; is a standard specific heat capacity of the soy milk at 25°C;

[0027] The main controller is further configured to obtain a final power duty ratio based on the initial power duty ratio and the initial heating control instruction. The initial heating control instruction is obtained to control the heating pipe.

[0028] Preferably, the main controller is configured to receive the liquid level of the soy milk in real time, and generate an initial trigger opening instruction when the liquid level of the soy milk is lower than the lower limit of the liquid level safety threshold range;

[0029] The current opening degree instruction of the water inlet electromagnetic valve is determined based on the historical actual opening instruction and the historical actual opening degree of the water inlet electromagnetic valve, and the water inlet electromagnetic valve is controlled to open the water replenishment.

[0030] When the liquid level data reaches the upper limit of the threshold range, an initial closing instruction is generated to control the water inlet electromagnetic valve to close.

[0031] Preferably, the main controller is configured to determine the upper limit and the lower limit of the liquid level safety threshold range according to the planned production set by the human-computer interaction interface for the current production batch;

[0032] , ;

[0033] wherein, is the upper limit; is the planned production based on the human-computer interaction interface; and k2 is a second empirical coefficient; is a centrifugal force correction coefficient, and , is a real-time rotating speed of the separation actuator, is a maximum rotating speed of the separation actuator; is a cross-sectional area of the separation bin; is a feed flow correction coefficient, and , is a real-time opening degree of the electromagnetic flow valve; is a maximum opening degree of the electromagnetic flow valve; is a full-bank height of the separation bin; is a comparison height value;

[0034]

[0035] wherein, is a lower limit value; is a hysteresis interval.

[0036] Preferably, the main controller comprises:

[0037] a calculation unit configured to calculate a difference function of the concentration fitting function and the turbidity fitting function, and perform cluster analysis on the concentration difference values at all collection time points according to the difference function to obtain a standard deviation, a mean value and a cluster weight of each cluster analysis result;

[0038] a first deviation unit configured to obtain a first sub-deviation set according to the standard deviation, the mean value and the cluster weight of each cluster analysis result, and in combination with a first constant of the concentration fitting function and a second constant of the turbidity fitting function;

[0039] a division unit configured to divide a function curve of the difference function according to a preset continuous collection time point to determine a single function of each division curve, and obtain a single constant of the single function;

[0040] a second deviation unit configured to perform discrete analysis on the single constant to determine whether there is a discrete constant, and obtain a second sub-deviation set according to a collection time point of the discrete constant and a mean value of a non-discrete constant;

[0041] a time group association unit configured to respectively sort the concentrations at all collection time points in the concentration fitting function and the turbidity fitting function in size, and determine a collection time group of a same concentration sorting serial number, and analyze a two-dimensional association relationship of the concentration fitting function and the turbidity fitting function according to all collection time groups;

[0042] an expansion unit configured to determine an optimal offset amount in combination with the two-dimensional association relationship and the first sub-deviation set and the second sub-deviation set, and expand a range of the initial vibration frequency in combination with a soybean milk liquid level height at a current time.

[0043] Preferably, the main controller further comprises:

[0044] a sequence acquisition unit configured to acquire a historical actual opening instruction sequence of the water inlet electromagnetic valve and a corresponding historical actual opening degree sequence and perform feature extraction to generate an opening degree time sequence feature vector Vh, wherein, , ​The j2th historical actual opening instruction and the historical actual opening degree, indicates opening, =0 indicates closing; mo is the total number of instructions in the sequence;

[0045] Constructing a real-time parameter vector , wherein, is the real-time liquid level value of the separation tank, is the real-time temperature of the soybean milk, is the target liquid level value;

[0046] A prediction unit is configured to input Vh and into a pre-trained opening degree prediction model to obtain the current opening degree instruction of the water inlet electromagnetic valve.

[0047] Preferably, the main controller is further configured to obtain a standard state of the automation system based on process parameters, and perform difference analysis on the working state of the automation system to obtain a difference vector;

[0048] The main controller is further configured to input the difference vector into a pre-trained vector analysis model to obtain a first adjustment factor of each initial motor control instruction and a second adjustment factor based on an initial heating control instruction, and obtain an adjusted motor control instruction and an adjusted heating control instruction to control the corresponding vibration motor driver and heating control circuit to work.

[0049] Compared with the prior art, the application has the following beneficial effects:

[0050] Through the main controller integrating the concentration self-adaptation, temperature-liquid level coordination and multi-stage control logic, combined with multi-sensor synchronous acquisition and actuator precise control, the pain points of parameter fragmentation control, manual intervention and long production change debugging in traditional soybean milk separation are completely solved.

[0051] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the written description and the accompanying drawings.

[0052] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0053] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0054] Figure 1A structural diagram of an automatic control system for soybean milk separation in an embodiment of the present application. DETAILED DESCRIPTION

[0055] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, in which it is understood that the preferred embodiments described below merely for the purpose of illustrating and explaining the present application, and are not intended to limit the present application.

[0056] The present application provides an automatic control system for soybean milk separation, which is applied to soybean product processing equipment, such as a soybean milk separation device. Figure 1 As shown in the figure, the automatic control system comprises:

[0057] A main controller, which is a core processing unit of the automatic control system, is internally provided with a pre-configured multi-stage control logic for soybean milk separation, wherein the multi-stage control logic comprises a concentration self-adaptive regulation stage and a temperature-liquid level coordinated control stage.

[0058] A sensor group, which is electrically connected to an input port of the main controller, is used to collect environmental parameters of the soybean milk separation process, and the sensor group at least comprises a liquid level sensor for detecting the soybean milk liquid level in a separation bin, a temperature sensor for detecting the real-time temperature of the soybean milk, a viscosity detection unit for indirectly inferring the concentration state of the soybean milk, and a photoelectric turbidity sensor for collecting the turbidity of the soybean milk before separation, wherein the viscosity detection unit and the photoelectric turbidity sensor are synchronously collected.

[0059] A man-machine interaction interface, which is in communication connection with the main controller, is used to receive the process parameters set by a user and to display the working state of the automatic control system in real time.

[0060] The main controller is used to analyze the environmental parameters to obtain an initial control instruction set, wherein the initial control instruction set comprises an initial motor control instruction for a vibration motor driver, an initial heating control instruction for a heating control circuit, and an initial trigger opening instruction or an initial closing instruction for a water inlet electromagnetic valve for the vibration motor driver.

[0061] The initial control instruction set is adjusted according to the process parameters and the working state, and is issued to an actuator group to perform corresponding actions, wherein the actuator group at least comprises:

[0062] A vibration motor driver, which is used to drive and control the vibration frequency and amplitude of a separation sieve.

[0063] A heating control circuit, which is used to adjust the power of a heating pipe according to the temperature feedback.

[0064] A water inlet electromagnetic valve, which is used to control the on-off of water supply to the separation bin.

[0065] In this embodiment, the soy product processing equipment is an industrialized equipment for soy product production. The scheme focuses on the matching equipment of the soy milk separation section, i.e., the special equipment for realizing the separation of soy milk and soy residue, such as a vibrating screen separator, a horizontal screw centrifugal separator, a negative pressure filter separator, etc. in a soy milk production line.

[0066] The main controller is selected from a programmable logic controller such as a Siemens S7-1200 CPU or a microcontroller such as an STM32.

[0067] The soy milk separation multi-stage control logic is an algorithm logic built in the main controller for realizing the accurate control of soy milk separation in stages, which solves the limitations of traditional single parameter control. The concentration self-adaptive regulation stage is a stage for automatically adjusting the vibration parameters according to the real-time concentration of soy milk to ensure the separation efficiency of soy milk residue. The temperature-liquid level cooperative control stage is a stage for correlating the real-time temperature of soy milk and the liquid level of the separation tank to avoid the waste of efficiency or quality problems caused by single regulation.

[0068] The liquid level sensor is selected from a submersible liquid level transmitter with a model of JYB-KO-Y2, a measurement range of 0-100 cm, and an output of 4-20 mA analog signal. The sensor probe is vertically submersed into the separation tank away from stirring or vibrating components. The cable is fixed on the tank top, and the signal cable is connected to the analog input terminal of the main controller.

[0069] The temperature sensor is selected from a PT100 platinum resistance temperature sensor such as OMRONE52-CA1D, with a measurement range of 0-150℃. The sensor probe is packaged in a food-grade stainless steel sleeve and inserted into the separation tank. The depth of the probe is to the middle of the soy milk. The resistance signal is converted into a 4-20 mA signal by a temperature transmitter and connected to the main controller.

[0070] The viscosity detection unit is selected from an online rotary viscometer such as Brookfield DV2T, with a measurement range of 10-10000 cP and an output of 4-20 mA analog signal. The measurement rotor of the viscometer is installed on the feed pipe of the separation tank and fixed through the pipe flange. The signal cable is connected to the main controller and synchronously triggered with the turbidity sensor.

[0071] The photoelectric turbidity sensor is selected from an online photoelectric turbidimeter such as HACH 2100Q, with a measurement range of 0-1000 NTU and an output of 4-20 mA analog signal. The transmitting end and the receiving end of the sensor are respectively installed on the two sides of the feed pipe and fixed through quick connectors. The signal is connected to the main controller and shares a trigger signal with the viscosity detection unit.

[0072] A timer is set in the main controller to output switch quantity trigger signals at a fixed time interval, which are connected to the external trigger interfaces of the viscometer and the turbidimeter, respectively. At the same time, a synchronous data receiving logic is set in the main controller program to ensure that the data of the two sensors are stored in the same data frame.

[0073] The human-computer interaction interface is the interactive medium between the user and the automatic control system, realizing the bidirectional function of process parameter input and system working state display. For example, a 7-inch or 10-inch industrial touch screen with a resolution of 1024x600 is selected, which is connected with the main controller through the RS485 communication interface.

[0074] The environmental parameters are the data set collected by the sensor group in real time, reflecting the current state of the soybean milk separation process, such as liquid level 75 cm, temperature 82℃, viscosity 480 cP, and turbidity 130 NTU.

[0075] The initial control instruction set is the instruction set generated by the main controller based on the environmental parameters and the preset basic logic for controlling the actuators, such as viscosity 480 cP and turbidity 130 NTU, the initial instruction is: vibration frequency 50 Hz, amplitude 2.0 mm.

[0076] The process parameters are the parameters set by the user through the human-computer interaction interface according to the production requirements, which are the basis for the main controller to adjust the initial instruction, such as planned output 1000 L, target temperature 85℃, reference vibration frequency of vibration motor driver 45 Hz, liquid level safety range 60-90 cm, and soybean seed correction coefficient 1.1.

[0077] The working state is the real-time running state data of the system, including the environmental parameters collected by the sensors and the current action state of the actuators, which is used by the main controller to determine whether the instruction needs to be adjusted.

[0078] The vibration motor driver selects a three-phase asynchronous vibration motor driver, such as Delta VFD022EL43A. The driver is installed in the electrical cabinet, the input end is connected with a 380V three-phase power supply, the output end is connected with the vibration motor, the control end is connected with the main controller through a signal line to receive the frequency and amplitude instructions, and the vibration motor is connected with the separation screen through a flange to ensure the vibration transmission efficiency.

[0079] The heating control circuit is composed of 220V AC power supply→ fuse→ solid-state relay→ heating tube→ thermal relay→ zero line. The control end of the solid-state relay is connected with the main controller, the circuit is installed on the insulating mounting plate in the electrical cabinet, and the heating tube is installed at the bottom or side wall of the separation bin. The thermal relay sets the overload protection current to prevent the heating tube from burning out.

[0080] The water inlet electromagnetic valve selects a two-position two-way food-grade electromagnetic valve, such as SMCVX212, with an interface specification of DN25. The electromagnetic valve is connected in series in the water replenishment pipeline, the water inlet end is connected with the tap water or purified water pipeline, and the water outlet end is connected with the top of the separation bin through a pipe clamp for fixation. The coil wiring end of the electromagnetic valve is connected with the on-off output terminal of the main controller to ensure firm connection and prevent loosening.

[0081] The beneficial effects of the above technical solutions are: through the main controller integrating the concentration self-adaptive, temperature-liquid level coordinated multi-stage control logic, combined with the multi-sensor synchronous acquisition and the precise control of the actuator, the pain points of the parameter fragmentation control, the manual intervention and the long change production debugging in the traditional soybean milk separation are completely solved; in actual application, the soybean milk separation efficiency can be improved by 5% to 8%, the product qualified rate fluctuation can be reduced from ±8% to ±2%, the change production debugging time can be shortened from 1 to 2 hours to 10 minutes, and 30% of the manual operation amount can be reduced, so that the high efficiency and stability requirements of the bean product industrialized continuous production are met.

[0082] The application provides an automatic control system for soybean milk separation, and the main controller is used for receiving a current signal collected by the viscosity detection unit as a first feedback value representing the concentration of soybean milk and receiving a current signal collected by the photoelectric turbidity sensor as a second feedback value representing the turbidity of soybean milk, and determining an initial vibration frequency of the vibration motor driver.

[0083] , and ;

[0084] Among them, The initial vibration frequency of the vibration motor driver is determined. The reference vibration frequency of the vibration motor driver is determined. The first feedback value is determined. The reference viscosity current value corresponding to the standard concentration is determined. The second feedback value is determined. The reference turbidity current value corresponding to the standard turbidity is determined. The viscosity influence weight coefficient is determined. The turbidity influence weight coefficient is determined. The separation stage correction coefficient is determined. The bean variety correction coefficient is determined. The frequency analysis function is determined.

[0085] The main controller is used for performing first comparison between the first feedback values obtained at different moments and a preset concentration-vibration parameter mapping table and performing second comparison between the second feedback values obtained at different moments and a preset turbidity-vibration parameter mapping table, and a two-dimensional array is constructed.

[0086] The main controller is used for performing fitting analysis on two rows of arrays in the two-dimensional array respectively, and performing ladder analysis on the concentration fitting function and the turbidity fitting function to obtain a set of offsets.

[0087] The main controller is used for expanding the range of the initial vibration frequency depending on the set of offsets, and screening a middle value in the expanded range as a target vibration frequency.

[0088] The main controller is configured to obtain a target vibration frequency based on the target vibration frequency obtain a target amplitude of the vibration motor driver;

[0089]

[0090] wherein, is an initial amplitude of the vibration motor driver; is a reference amplitude of the vibration motor driver; is an influence coefficient of turbidity on amplitude; is a frequency correction term; is a maximum vibration frequency in the two-dimensional array; is a mesh aperture; is a vibration motor shaft distance length;

[0091] The main controller is configured to obtain an initial motor control instruction for controlling the vibration motor driver to operate based on the target vibration frequency and the target amplitude.

[0092] In this embodiment, the optimal frequency under different working conditions is tested in a laboratory, and the frequency under a standard working condition is stored in a parameter table of the main controller.

[0093] In this embodiment, for example, soybean milk with a concentration of 8%, is 10 mA, and when turbidity is 100 NTU, is 8 mA; is a coefficient for measuring the influence of viscosity on vibration frequency, and the greater the value, the more significant the influence, wherein the influence of viscosity on frequency is greater than that of turbidity, and the coefficient is obtained according to a large number of experiments is 0.8, and + . According to the coefficients adjusted in the separation stages, i.e., the initial stage, the middle stage, and the late stage of separation, the change rules of the amount of soybean dregs and the concentration are different in different stages. Generally, in the initial stage of separation, i.e., the first 10 minutes, is 1, 2, high frequency is required to prevent clogging, in the middle stage of separation is 1.0, and in the late stage of separation is 0.9.

[0094] The soybean species correction coefficient is also different according to different soybeans, such as yellow beans, red beans, and black beans. For example, of yellow beans is 1.0, that of black beans is 1.1 because the particles are hard, and that of red beans is 0.9 because the particles are small.

[0095] In this embodiment, the concentration-vibration parameter mapping table is a pre-established parameter table of soybean milk concentration-optimal vibration frequency, such as 10 mA of viscosity current corresponding to 8% of concentration, and at this time, the frequency is 50 Hz, etc. ​

[0096] In this embodiment, the turbidity-vibration parameter mapping table is a soybean milk turbidity-optimal vibration frequency parameter table established in advance through experiments, for example, when the turbidity current is 8, i.e., the turbidity is 100 NTU, the frequency is 50 Hz.

[0097] The two-dimensional array is a time sequence two-dimensional data structure of concentration comparison results and turbidity comparison results at different times, for example, the concentration at time t1 corresponds to the frequency 55 Hz, and the turbidity corresponds to the frequency 53 Hz, and at this time, the two-dimensional data is: .

[0098] In this embodiment, the fitting analysis is used to approximate the change trend of the time-vibration parameter in the two-dimensional array, and the concentration and turbidity fitting functions are obtained, and the fitting functions are linear functions: y=kx+b.

[0099] For example, the concentration fitting function is y1=k01t+b1, wherein k01 is 0.5 Hz / minute, and b1 is 50 Hz, and the turbidity fitting function is y2=k02t+b2, wherein k02 is 0.3 Hz / minute, and b2 is 52 Hz.

[0100] In this embodiment, the step analysis is to divide the difference function of the fitting function according to the preset continuous acquisition time, and the preset continuous acquisition time can be M0 continuous time, and then the offset set is obtained, and the range expansion is realized.

[0101] In this embodiment, the target vibration frequency is the average value calculated from the upper and lower limits of the expanded range.

[0102] In this embodiment, the reference amplitude is the reference amplitude of the vibration motor under the standard soybean, the standard concentration / turbidity, and the influence coefficient of turbidity on amplitude is obtained according to a large number of fitting calculations.

[0103] In this embodiment, the screen aperture is directly obtained according to the screening model, and the vibration motor shaft distance length is determined when the equipment is shipped, and can be directly used.

[0104] The beneficial effects of the above technical scheme are: through the synchronous feedback of the viscosity and turbidity sensors, combined with the formula calculation of the soybean, the separation stage and other correction coefficients and the dynamic adjustment of the two-dimensional array fitting and step analysis, the precise collaborative control of the vibration motor frequency and amplitude is realized; the slurry residue separation efficiency is improved, the screen clogging probability is reduced, and the continuity and stability of the soybean milk separation process are ensured.

[0105] The application provides an automatic control system for soybean milk separation, wherein the main controller is used to calculate the output power duty cycle of the heating control circuit.

[0106] ;

[0107] in, The duty cycle of the real-time output power of the heating control circuit; This is the proportionality coefficient; Temperature deviation; Let be the temperature segmentation coefficient, and ; The integral time constant; This is the integral correction factor, and ; The differential time constant; Forward compensation coefficient; The target temperature at the current moment in the preset temperature curve; The real-time temperature of the soy milk detected by the temperature sensor; This represents the real-time density of the soy milk at the current temperature. This represents the real-time specific heat capacity of soy milk at the current temperature. This is the standard density of soy milk at 25℃; This refers to the standard specific heat capacity of soy milk at 25℃.

[0108] The main controller is also used to base the final power duty cycle. The initial heating control command for the heating element is obtained.

[0109] In this embodiment, the preset temperature curve is a pre-set target temperature trajectory that changes over time, guiding the temperature change pattern of soy milk heating.

[0110] In this embodiment, the feedforward compensation coefficient is used to compensate for temperature change trends in advance, for example... It is 1.5, when When the temperature is 2℃ / minute, the output of the feedforward circuit is 1.5×2=3.

[0111] The real-time density of soy milk is based on Found from a temperature-density comparison table, for example, At 80℃, for .

[0112] In this embodiment, the main controller is based on The generated commands control the on / off state of the heating element or adjust the power, for example, At this time, a PWM signal with a period of 1 second and a high level of 0.6 seconds is output to control the heating tube to work at 60% power.

[0113] The beneficial effects of the above technical scheme are: the improved PID control algorithm is adopted, the temperature segmented coefficient and the integral correction coefficient are combined to adapt to the control characteristics of different temperature sections, the temperature change trend is responded in advance through feedforward compensation, and the range of the power duty cycle is restricted, compared with the traditional heating control, the temperature control precision of soybean milk heating is improved, the heating efficiency is improved, the damage of the heating tube or the temperature out of control caused by power over-limit is avoided, and the stability of the soybean milk separation temperature condition is ensured.

[0114] The application provides an automatic control system for soybean milk separation, wherein the main controller is used for receiving the soybean milk liquid level height collected by the liquid level sensor in real time, and generating an initial trigger opening instruction when the soybean milk liquid level height is lower than the lower limit value of the liquid level safety threshold range.

[0115] The current opening degree instruction of the water inlet electromagnetic valve is determined according to the historical actual opening instruction and the historical actual opening degree of the water inlet electromagnetic valve, and the water inlet electromagnetic valve is controlled to open and supplement water.

[0116] When the liquid level data reaches the upper limit of the threshold range, an initial closing instruction is generated to control the water inlet electromagnetic valve to close.

[0117] Preferably, the main controller is used for determining the upper limit value and the lower limit value of the liquid level safety threshold range required by the current production batch according to the planned yield set by the man-machine interaction interface.

[0118]

[0119] wherein, is the upper limit value; is the planned yield set by the man-machine interaction interface; k2 is the second empirical coefficient; is the centrifugal force correction coefficient, and , is the real-time rotating speed of the separation actuator, is the maximum rotating speed of the separation actuator; is the cross-sectional area of the separation bin; is the feed flow correction coefficient, and , is the real-time opening degree of the electromagnetic flow valve; is the maximum opening degree of the electromagnetic flow valve; is the full-bin height of the separation bin; is the comparison height value;

[0120]

[0121] wherein, is the lower limit value; is the hysteresis interval. ​​​

[0122] In this embodiment, n = 1500 r / min, For 2000 r / min, then = 1.0375, = 50%, For 100%, then For 1.015, for example, the maximum holding height of the separation tank is 100 cm.

[0123] The hysteresis interval prevents the valve from frequently starting and stopping, for example, = 10 cm.

[0124] Preferably, the main controller further comprises:

[0125] a sequence acquisition unit for acquiring a historical actual opening instruction sequence of the water inlet electromagnetic valve and a corresponding historical actual opening degree sequence and performing feature extraction to generate an opening degree time sequence feature vector Vh, wherein, 、 are the j2th historical actual opening instruction and historical actual opening degree, respectively, indicates opening, = 0 indicates closing; mo is the total number of instructions in the sequence;

[0126] a real-time parameter vector is constructed , wherein, is the real-time liquid level value of the separation tank, is the real-time temperature of the soybean milk, is the target liquid level value;

[0127] a prediction unit for inputting Vh and into a pre-trained opening degree prediction model to obtain the current opening degree instruction of the water inlet electromagnetic valve.

[0128] In this embodiment, the initial trigger opening instruction is when the liquid level is below the lower limit, the main controller generates an initial signal to start the water replenishment.

[0129] The historical actual opening instruction of the water inlet electromagnetic valve is the opening (1) / closing (0) instruction sequence sent by the main controller in the past, such as {1, 0, 1}, indicating the first opening, the second closing, and the third opening. The historical actual opening degree is the actual opening degree of the electromagnetic valve corresponding to the opening instruction, for example, {60%, 0%, 70%}.

[0130] In this embodiment, the current opening degree instruction is the degree that the main controller calculates according to the historical instruction and opening degree, such as the opening degree of 65% when the historical liquid level is 45 cm, and the current instruction of 65% is generated.

[0131] In this embodiment, the second empirical coefficient is a correction coefficient determined by experiment, for example, the k2 of soybean milk is 1.2.

[0132] In this embodiment, the opening time sequence feature vector Vh is an ordered numerical group organized in a fixed dimension, which converts the unstructured historical sequence into a structured input for the pre-trained opening prediction model to learn the historical law and thus predict the opening of the current electromagnetic valve, for example, Vh can be expressed as:

[0133] .

[0134] In this embodiment, the pre-trained opening prediction model is a machine learning model trained based on more than 1000 groups of historical data, the input of which is Vh and , and the output is the current opening instruction.

[0135] The beneficial effects of the above technical solution are: through real-time acquisition of the liquid level by the main controller and combination of historical data, the liquid level safety threshold is dynamically calculated, and the pre-trained model is used to accurately predict the opening of the electromagnetic valve, solving the problem of large fluctuation and poor adaptability of the traditional fixed threshold liquid level, and the liquid level control precision is within ±2cm, the water supply response speed is accelerated by 40%, manual debugging is reduced, the soybean milk separation liquid level is stabilized, and the slurry separation efficiency and product quality consistency are indirectly improved.

[0136] The application provides an automatic control system for soybean milk separation, the main controller comprising:

[0137] The calculation unit is configured to calculate a difference function of the concentration fitting function and the turbidity fitting function, and perform clustering analysis on the concentration difference values at all acquisition time points according to the difference function to obtain a standard deviation, a mean value and a clustering weight of each clustering analysis result.

[0138] The first deviation unit is configured to obtain a first sub-deviation set according to the standard deviation, the mean value and the clustering weight of each clustering analysis result, and in combination with a first constant of the concentration fitting function and a second constant of the turbidity fitting function.

[0139] The division unit is configured to divide the function curve of the difference function according to the preset continuous acquisition time points to determine a single function of each division curve, and obtain a single constant of the single function.

[0140] The second deviation unit is configured to perform discrete analysis on the single constant to determine whether there is a discrete constant, and obtain a second sub-deviation set according to the acquisition time of the discrete constant and the average value of the non-discrete constant.

[0141] The time group association unit is configured to sort the concentrations at all collection time points in the concentration fitting function and the turbidity fitting function in descending order respectively, determine a collection time group with the same concentration sorting sequence number, and analyze the two-dimensional correlation between the concentration fitting function and the turbidity fitting function according to all collection time groups.

[0142] The expansion unit is configured to determine the optimal offset based on the first sub-deviation set and the second sub-deviation set and in combination with the two-dimensional correlation, and expand the range of the initial vibration frequency in combination with the soybean milk liquid level at the current time.

[0143] In this embodiment, the difference function is fitted based on the concentration difference between the concentration fitting function and the turbidity fitting function at the same collection time point. At this time, the standard deviation, mean value, and clustering weight of each clustering result are obtained by clustering analysis of all concentration differences, wherein the clustering weight is the proportion of the data amount in the clustering result to the total data amount, and the clustering algorithm is implemented by using the K-means algorithm. For example, the K-means algorithm clusters the differences into two groups: {0, 1} and {2, 3, 4}. At this time, the mean value of the first group is 0.5, the standard deviation is 0.5, and the clustering weight is 0.4. The mean value of the second group is 3, the standard deviation is 1, and the clustering weight is 0.6.

[0144] In this embodiment, the first constant and the second constant are intercepts of the corresponding functions.

[0145] At this time, the deviation of each clustering result is: (the mean value + the standard deviation of the corresponding clustering result) × the corresponding weight + (the first constant - the second constant), and the first sub-deviation set: {the deviation of each clustering result} is obtained.

[0146] In this embodiment, the preset continuous collection time points are set in advance in units of 3, and the single function is a linear function fitted once, and the single constant is the intercept of the corresponding single function.

[0147] In this embodiment, the discrete analysis is performed by using 3 The principle is used to determine whether it is a discrete constant. For example, the first constant set is {1, 1, 5, 1}. By using 3 The principle identifies that 5 is a discrete constant, the time of the discrete constant is time 3, the average value of the non-discrete constant is 1, the second sub-deviation set {0, 0, 4, 0} is obtained by calculating the deviation, wherein the deviation of the discrete constant is 5-1=4, and the deviation of the non-discrete constant is 0.

[0148] In this embodiment, the collection time groups with the same concentration sorting sequence number are, for example, {the collection time corresponding to the first concentration and the first turbidity forms a group}.

[0149] For example, the concentration fitting values at time points 1, 2, 3, 4, and 5 are 3, 5, 7, 9, and 11 respectively, at this time, the sequence number 1 corresponds to the time point 5 after sorting, the sequence number 2 corresponds to the time point 4, and so on. The turbidity fitting values at time points 1, 2, 3, 4, and 5 correspond to 3, 4, 5, 6, and 7 respectively, the sequence number 1 corresponds to the time point 5 after sorting, the sequence number 2 corresponds to the time point 4, and so on. The collection time groups with the same sequence number are (5, 5), (4, 4), and so on. The two-dimensional correlation is positively correlated with the concentration and the turbidity, and the sorting sequence numbers are consistent. The two-dimensional correlation is determined by calculating the correlation coefficient.

[0150] In this embodiment, the Pearson correlation coefficient r is calculated based on the concentration fitting value and the turbidity fitting value in all collection time groups according to the Pearson correlation coefficient. It is a known attempt. If r is greater than 0, it indicates a positive correlation. If r is less than 0, it indicates a negative correlation. The closer r is to 1 or -1, the stronger the correlation.

[0151] The sorting sequence number, the corresponding concentration value, and the turbidity value of each time group are extracted, and the sequence number-concentration curve and the sequence number-turbidity curve are drawn respectively. If the rising, falling, or stable trends of the two curves are synchronized, it is determined that the trends are consistent. Otherwise, it is determined that the trends are opposite.

[0152] The concentration values and the turbidity values are divided into several intervals according to the equidistant interval respectively, and the co-occurrence frequency of each concentration interval-turbidity interval is counted.

[0153] If the co-occurrence frequency of the high concentration interval and the high turbidity interval is high, and the co-occurrence frequency of the low concentration interval and the low turbidity interval is high, it indicates that the distribution matching degree is high.

[0154] If the co-occurrence frequency of the high concentration interval and the low turbidity interval is high, it indicates that the distribution matching degree is low.

[0155] For example: The high concentration {10%, 12%} and the high turbidity {25, 30} co-occur twice, which is the main co-occurrence combination, indicating that the distribution matching degree is high.

[0156] The results of the Pearson correlation coefficient, the trend consistency, and the distribution matching degree are combined to comprehensively determine the type of the correlation:

[0157] If r≥0.8, the trend consistency≥90%, and the distribution matching degree is high, at this time, it is determined that the correlation is highly synchronous. If r≥0.8, the trend consistency≤30%, and the distribution matching degree is low, at this time, it is determined that the correlation is weak / irrelevant.

[0158] If r<−0.8, the trend consistency≥90%, but the trends are opposite, and the high concentration corresponds to the low turbidity and the low concentration corresponds to the high turbidity, at this time, it is determined that the correlation is highly reverse.

[0159] If r<−0.8, the trend consistency≤30%, and the distribution matching degree is low, at this time, it is determined that the correlation is weak / irrelevant. ​

[0160] Otherwise, it is determined as a general association.

[0161] In this embodiment, the mean values of the first and second sub-deviation sets are D1 and D2 respectively, and the comprehensive influence amount f01=D1×wu1+D2×wu2 is calculated by combining the weights wu1 and wu2 of the two-dimensional association.

[0162] The experience offset is directly obtained according to the industrial control experience, and f02 is +2Hz.

[0163] According to the liquid level deviation, the expanded basis f03=(ideal liquid level height-bean milk liquid level height)×0.5Hz / cm is calculated.

[0164] The expanded vibration frequency range is [ft-f03+f01+f02, ft+f03+f01+f02].

[0165] The beneficial effects of the above technical solution are: through multi-unit cooperation, difference analysis, clustering, deviation calculation, segmented fitting and two-dimensional association analysis of the fitting function of concentration and turbidity, and finally combining the dynamic expansion of the vibration frequency range with the liquid level, the limitation of poor adaptability of vibration parameters in traditional control is broken, the dynamic changes of the concentration, turbidity and liquid level of the bean milk are adaptively adjusted, the vibration frequency is more accurately matched with the slurry residue separation demand, and the separation efficiency and the stability of the bean milk quality are improved.

[0166] The application provides an automatic control system for bean milk separation, and the main controller is also used for obtaining a standard state of the automatic system based on process parameters, and performing difference analysis on a working state of the automatic system to obtain a difference vector.

[0167] The main controller is also used for inputting the difference vector into a pre-trained vector analysis model to obtain a first adjustment factor of each initial motor control instruction and a second adjustment factor based on an initial heating control instruction, and obtaining an adjusted motor control instruction and an adjusted heating control instruction to control a corresponding vibration motor driver and heating control circuit to work.

[0168] In this embodiment, the standard state is an ideal running state of the system when the process parameters are completely met, and is obtained from a pre-set process parameter-standard state mapping table, for example, the process parameters are 98 DEG C, 45Hz, the heating power empty occupation ratio is 80% in the standard state, the vibration frequency is stable at 45Hz, and the liquid level is 60cm; the working state is a real-time state, for example, the actual temperature is 92 DEG C, the vibration frequency is 42Hz, and the liquid level is 55cm, at this time, the difference vector is: {-6 DEG C, -3Hz, -5cm}.

[0169] In this embodiment, the pre-trained vector analysis model is a machine learning model trained by a large number of difference vectors-regulating factors data, for example, the output difference vector is {-6℃, -3Hz, -5cm}, the output first regulating factor is +5Hz, and the second regulating factor is 1.2 times, at this time, the coefficient of the adjusted initial heating instruction is: 70%×1.2, wherein 70% is the heating power proportion corresponding to the initial instruction.

[0170] The beneficial effects of the above technical solutions are: through standard-actual state difference analysis and intelligent adjustment combined with a pre-trained model, the soybean milk separation working condition fluctuation can be adapted in real time, and the separation efficiency and quality stability can be ensured.

[0171] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. An automatic control system for soy milk separation, applied to a bean product processing device, characterized in that, The application relates to an automatic control system for a soybean milk separation device, which comprises the following components: a main controller, which is a core processing unit of the automatic control system and is internally provided with preconfigured multi-stage control logic for soybean milk separation, wherein the multi-stage control logic comprises a concentration self-adaptive regulation stage and a temperature-liquid level cooperative control stage; a sensor group, which is electrically connected to an input port of the main controller and is used for collecting environmental parameters of a soybean milk separation process, wherein the sensor group at least comprises a liquid level sensor for detecting the liquid level of soybean milk in a separation tank, a temperature sensor for detecting the real-time temperature of the soybean milk, a viscosity detection unit for indirectly inferring the concentration state of the soybean milk, and a photoelectric turbidity sensor for collecting the turbidity of the soybean milk before separation, wherein the viscosity detection unit and the photoelectric turbidity sensor are used for synchronous collection; a man-machine interactive interface, which is in communication connection with the main controller and is used for receiving process parameters set by a user and displaying the working state of the automatic control system in real time; the main controller, which is used for analyzing the environmental parameters to obtain an initial control instruction set, wherein the initial control instruction set comprises an initial motor control instruction for a vibrating motor driver, an initial heating control instruction for a heating control circuit and an initial trigger opening instruction or an initial closing instruction for a water inlet electromagnetic valve of the vibrating motor driver; the initial control instruction set is adjusted according to the process parameters and the working state, and is sent to an executor group to execute corresponding actions, wherein the executor group at least comprises: a vibrating motor driver, which is used for driving and controlling the vibration frequency and amplitude of a separation screen; a heating control circuit, which is used for adjusting the power of a heating pipe according to the temperature feedback; a water inlet electromagnetic valve, which is used for controlling the on-off of water supplement to the separation tank; wherein the main controller is used for receiving the current signal collected by the viscosity detection unit as a first feedback value representing the concentration of the soybean milk and receiving the current signal collected by the photoelectric turbidity sensor as a second feedback value representing the turbidity of the soybean milk, and determining the initial vibration frequency of the vibrating motor driver; , and ; wherein, is an initial vibration frequency of the vibration motor driver; is a reference vibration frequency of the vibration motor driver; is a first feedback value; is a reference viscosity current value corresponding to a standard concentration; is a second feedback value; is a reference turbidity current value corresponding to a standard turbidity; is a viscosity influence weight coefficient; is a turbidity influence weight coefficient; is a separation stage correction coefficient; is a bean species correction coefficient; is a frequency analysis function; the main controller is used for performing first comparison between the first feedback value obtained at different moments and a preset concentration-vibration parameter mapping table and performing second comparison between the second feedback value obtained at different moments and a preset turbidity-vibration parameter mapping table, and constructing a two-dimensional array; the main controller is used for performing fitting analysis on two rows of arrays in the two-dimensional array respectively, and performing ladder analysis on a concentration fitting function and a turbidity fitting function to obtain a set of offsets; the main controller is used for expanding the range of the initial vibration frequency according to the set of offsets, and screening a middle value in the expanded range as a target vibration frequency; The main controller is configured to obtain the target vibration frequency based on the target amplitude of the vibration motor driver obtain the target amplitude of the vibration motor driver ； wherein, is an initial amplitude of the vibration motor driver; is a reference amplitude of the vibration motor driver; is a turbidity-to-amplitude influence coefficient; is a frequency correction term; is a maximum vibration frequency in the two-dimensional array; is a screen mesh aperture; is a vibration motor axle distance length; the main controller is used for obtaining an initial motor control instruction for controlling the vibrating motor driver to work based on the target vibration frequency and a target amplitude; wherein the main controller comprises: a calculation unit, which is used for calculating the difference function of the concentration fitting function and the turbidity fitting function, and performing clustering analysis on the concentration difference values at all collection moments according to the difference function to obtain the standard deviation, the mean value and the clustering weight of each clustering analysis result. The first deviation unit is configured to obtain a first sub-deviation set according to a standard deviation, a mean value and a clustering weight of each clustering analysis result, and in combination with a first constant of the concentration fitting function and a second constant of the turbidity fitting function; The division unit is configured to divide the function curve of the difference function according to the preset continuous collection time to determine a single function of each divided curve, and obtain a single constant of the single function; The second deviation unit is configured to perform discrete analysis on the single constant to determine whether there is a discrete constant, and obtain a second sub-deviation set according to a collection time of the discrete constant and a mean value of the non-discrete constant; The time group association unit is configured to sort the concentrations in the concentration fitting function and the turbidity fitting function in all collection times in size, respectively, determine a collection time group of a same concentration sorting sequence number, and analyze a two-dimensional association relationship of the concentration fitting function and the turbidity fitting function according to all collection time groups; The expansion unit is configured to determine an optimal offset in combination with the first sub-deviation set and the second sub-deviation set and the two-dimensional association relationship, and expand the range of the initial vibration frequency in combination with the soybean milk liquid level at the current time.

2. The automated control system for soy milk separation according to claim 1, wherein, The main controller is configured to calculate an output power duty cycle of the heating control circuit; ； wherein, is a real-time output power duty cycle of the heating control circuit; is a proportional coefficient; is a temperature deviation; is a temperature segment coefficient, and ; is an integral time constant; is an integral correction coefficient, and ; is a differential time constant; is a feedforward compensation coefficient; is a target temperature at a current time in a preset temperature curve; is a real-time temperature of the soy milk detected by the temperature sensor; is a real-time density of the soy milk at a current temperature; is a real-time specific heat capacity of the soy milk at a current temperature; is a standard density of the soy milk at 25°C; is a standard specific heat capacity of the soy milk at 25°C; The main controller is further configured to determine the final power duty cycle based on the initial heating control instruction An initial heating control instruction for controlling the heating tube is obtained.

3. The automated control system for soy milk separation according to claim 2, wherein, The main controller is configured to receive the soybean milk liquid level collected by the liquid level sensor in real time, and generate an initial trigger opening instruction when the soybean milk liquid level is lower than a lower limit value of a liquid level safety threshold range; The main controller is configured to determine a current opening degree instruction of the water inlet electromagnetic valve in combination with a historical actual opening instruction and a historical actual opening degree of the water inlet electromagnetic valve, and control the water inlet electromagnetic valve to open and supplement water; The main controller is configured to generate an initial closing instruction to control the water inlet electromagnetic valve to close when the liquid level data reaches an upper limit of the threshold range.

4. The automated control system for soy milk separation according to claim 3, wherein, The main controller is configured to determine an upper limit value and a lower limit value of the liquid level safety threshold range required for a current production batch according to a planned yield set by the human-computer interaction interface; ， ； wherein, is an upper limit value; is a planned production based on the human-machine interface setting; k2 is a second empirical coefficient; is a centrifugal force correction coefficient, and , is a real-time rotational speed of the separation actuator, is a maximum rotational speed of the separation actuator; is a cross-sectional area of the separation bin; is a feed flow correction coefficient, and , is a real-time opening degree of the electromagnetic flow valve; is a maximum opening degree of the electromagnetic flow valve; is a full-bin height of the separation bin; is a comparison height value; ； wherein is a lower limit value; is a hysteresis interval.

5. The automated control system for soy milk separation according to claim 4, wherein, The main controller further comprises: a sequence acquisition unit configured to acquire a historical actual opening instruction sequence of the water inlet electromagnetic valve and a corresponding historical actual opening degree sequence and perform feature extraction to generate an opening degree time sequence feature vector Vh, wherein, , are the j2th historical actual opening instruction and historical actual opening degree, respectively, represents opening, =0 represents closing; mo is the total number of instructions in the sequence; constructing a real-time parameter vector wherein, is a real-time liquid level value of the separation tank, is a real-time temperature of the soy milk, is a target liquid level value; a prediction unit configured to predict Vh based on the inputted Vh and inputting the Vh into the pre-trained opening degree prediction model to obtain an output current opening degree instruction of the water inlet electromagnetic valve.

6. The automated control system for soy milk separation of claim 1, wherein, The main controller is further configured to obtain a standard state of the automatic control system based on process parameters, and perform difference analysis on a working state of the automatic control system to obtain a difference vector; The main controller is further configured to input the difference vector into a pre-trained vector analysis model to obtain a first adjustment factor of each initial motor control instruction and a second adjustment factor based on an initial heating control instruction, and obtain an adjusted motor control instruction and an adjusted heating control instruction to control the corresponding vibration motor driver and heating control circuit to work.

Images