Control device, food processing machinery, and control method

Abstract

To provide a control device and food processing machinery that can improve the accuracy of liquid temperature control. [Solution] The control device 1 acquires parameters related to the liquid temperature and adjusts the operating amount of each control unit of the food processing machine 100 based on the mutual influence of these parameters. In this case, the control device 1 can adjust the operating amount of each control unit while considering the influence of other temperature-related parameters (such as flow rate and back pressure) in addition to the liquid temperature. Therefore, the control device 1 can control each control unit more appropriately than when only the liquid temperature is considered. As a result, the accuracy of liquid temperature control can be improved.

Description

Technical Field

[0001] The present disclosure relates to a control device, a food machine, and a control method.

Background Art

[0002] As a food machine for liquid food in the prior art, the one described in Patent Document 1 is known. This food machine for liquid food has a heat exchanger and heats a liquid such as liquid food in the heat exchanger.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Here, in the above-described food machine, a control device for performing stable temperature control is applied. For example, the control device has performed PID control for calculating the control amount of the control valve of the heat medium. However, the PID control may be affected by disturbances such as flow rate fluctuations, resulting in a decrease in the accuracy of temperature control.

[0005] An object of the present disclosure is to provide a control device and a food machine that can improve the accuracy of temperature control of a liquid or liquid food.

Means for Solving the Problems

[0006] The control device according to the present disclosure is a control device that controls the temperature of a liquid or liquid food in a food machine for liquid food, acquires parameters related to the temperature of the liquid or liquid food, and adjusts the operation amount of each operation unit of the food machine based on the mutual influence of the parameters.

[0007] The control device acquires parameters related to the temperature of the liquid or liquid food and adjusts the operating volume of each control unit of the food processing machine based on the interaction of these parameters. In this case, the control device can adjust the operating volume of each control unit while considering the influence of other temperature-related parameters (such as flow rate and back pressure), not just the temperature of the liquid or liquid food. Therefore, the control device can control each control unit more appropriately than when only the temperature of the liquid or liquid food is considered. As a result, the accuracy of temperature control of the liquid or liquid food can be improved.

[0008] The control device may modify the manipulated parameters while acquiring them during the operation of the food processing machine. In this case, the control device can perform accurate temperature control in real time while the food processing machine is in operation.

[0009] The control device may predict the state of the food processing machine after a finite length of time has elapsed by acquiring parameters, and modify the manipulated variable based on the prediction result. In this case, the control device can anticipate the state of the food processing machine and take preventative measures before the liquid or liquid food deviates from the appropriate temperature.

[0010] The control device may use a predictive model to control the temperature of liquids or liquid foods in a food processing machine. In this case, the control device can easily and accurately control the temperature by applying the observed values ​​of acquired parameters and the manipulated quantities for each control unit to the predictive model.

[0011] You may train the predictive model at least once a day. This will help keep the predictive model in an optimal state.

[0012] The control device may prepare multiple predictive models and switch between them based on the conditions in the food processing machine. In this case, the control device can perform temperature control with high accuracy according to the conditions of the food processing machine.

[0013] During maintenance or line modification of the food machine, the control device may perform re-learning of the prediction model. In this case, the control device can perform accurate temperature control in consideration of the effects of maintenance and line modification.

[0014] The food machine for liquid food is equipped with the above-mentioned control device.

[0015] The control method is a control method for controlling the temperature of a liquid or liquid food in a food machine for liquid food, which acquires parameters related to the temperature of the liquid or liquid food and adjusts the operation amount of each operation part of the food machine based on the mutual influence of the parameters.

[0016] According to the food machine and the control method, the same effects as those of the above-mentioned control device can be obtained.

Effect of the Invention

[0017] According to the present disclosure, it is possible to provide a control device and a food machine that can improve the accuracy of temperature control of a liquid or liquid food.

Brief Description of the Drawings

[0018] [Figure 1] It is a block diagram of a continuous sterilization device. [Figure 2] It is a system diagram showing the configuration of the continuous sterilization device in more detail. [Figure 3] It is a diagram showing the mutual influence of parameters and the content of adjustment of the operation amount based thereon. [Figure 4] It is a diagram showing the block configuration of the control unit. [Figure 5] It is a flowchart showing the content of the learning process. [Figure 6] It is a flowchart showing the content of the processing of the control unit. <I [Figure 7] It is a system diagram showing the continuous sterilization device according to the comparative example. [Figure 8] It is a system diagram showing a CIP device. [Figure 9]It is a system diagram showing a food machine according to a modified example. [Figure 10] It is a system diagram showing a food machine according to a modified example.

Embodiments for Carrying Out the Invention

[0019] Hereinafter, embodiments of a control device 1 for controlling the temperature of a liquid or liquid food in a food machine 100 according to the present disclosure will be described with reference to the drawings. In the description of the drawings, the same or corresponding elements are denoted by the same reference numerals, and redundant descriptions are appropriately omitted.

[0020] First, referring to FIG. 1, a food machine 100 that is a control target of the control device 1 will be described. The control device 1 is a device that controls the temperature of a liquid or liquid food in the food machine 100 for liquid food. The food machine 100 includes a food manufacturing device, food manufacturing equipment, and the like. In the present embodiment, a continuous sterilization device 200 is adopted as the food machine 100. The continuous sterilization device 200 is a device that sterilizes liquid food. The liquid food is not particularly limited, and examples thereof include drinking water, confectionery, seasonings, and the like. Specifically, examples of the liquid food include tea, coffee, juice, yogurt, jelly, jam, dashi, mentuyu, dressing (emulsified liquid dressing, semi-solid dressing, etc.), tare or sauce or beverage containing solids (sesame, sawdust, fiber, etc.), cream, and the like. FIG. 1 is a block diagram of the continuous sterilization device 200. FIG. 2 is a system diagram showing the configuration of the continuous sterilization device 200 in more detail.

[0021] As shown in FIG. 1, the continuous sterilization device 200 includes a balance tank 2, a liquid feed pump 3, a heat exchanger 4 for heating, a holding tube 6, a heat exchanger 7 for cooling, a back pressure regulator 8, a flow rate sensor 10, a temperature sensor 11, a temperature sensor 12, a pressure sensor 13, and a control device 1.

[0022] Balance tank 2 stores liquid food and functions as a source of liquid food. Balance tank 2 temporarily stores the liquid food to be processed and supplies a stable flow rate to the entire apparatus. Liquid pump 3 takes liquid food from balance tank 2 and sends it to the downstream processing unit. Liquid pump 3 controls the flow rate of liquid food and supplies it continuously and at a constant rate. Liquid pump 3 operates based on the control value received from the control unit 20 of the control device 1. Liquid pump 3 adjusts the flow rate of liquid food based on the control value. More specifically, liquid pump 3 operates (adjusts the control amount) based on observations from the flow sensor 10 and plays a role in maintaining a constant flow rate throughout the entire apparatus.

[0023] The heating heat exchanger 4 is a device for heating liquid food to a predetermined temperature. The heating heat exchanger 4 exchanges heat between the liquid food and a heating medium (e.g., steam or hot water) to raise the temperature of the liquid food. The heating heat exchanger 4 heats the liquid food to an appropriate temperature in the initial stages of the sterilization process. The heating medium supply unit 30, which supplies the heating medium to the heating heat exchanger 4, opens the control valve 31 to supply the heating medium from the supply source to the circulation system 32, and the pump 33 circulates the heating medium in the circulation system (see Figure 2). The control valve 31 operates based on the operation value received from the control unit 20 of the control device 1. The temperature of the heating medium in the circulation system 32 is adjusted by the opening degree of the control valve 31.

[0024] The holding tube 6 is a tube for holding heated liquid food for a certain period of time. The holding tube 6 maintains the liquid food at a predetermined temperature within the tube and holds it for the time necessary to ensure a sterilization effect. The holding tube 6 is composed of a tube of a predetermined length (see Figure 2).

[0025] The cooling heat exchanger 7 is a device for cooling the liquid food after it has passed through the holding tube 6. The cooling heat exchanger 7 exchanges heat between the liquid food and the cooling medium (e.g., cold water or cooling gas) to lower the temperature of the liquid food (see Figure 2). The cooling heat exchanger 7 cools the sterilized liquid food to an appropriate temperature. The control valve 35 operates based on the operating value received from the control unit 20 of the control device 1. The flow rate of the cooling medium is adjusted by the opening of the control valve 35. Note that heat exchangers 4 and 7 may be plate heat exchangers or tube heat exchangers.

[0026] The back pressure regulator 8 is a device for maintaining a constant pressure within the apparatus. The back pressure regulator 8 adjusts the resistance to the flow of liquid food and stabilizes the overall pressure of the apparatus. The back pressure regulator 8 is composed of a control valve (see Figure 2). The liquid food that has passed through the back pressure regulator 8 is supplied to the next process tank 34 for the next process. The back pressure regulator 8 operates based on the operating value received from the control unit 20 of the control device 1. The back pressure regulator 8 adjusts the pressure of the liquid food based on the operating value.

[0027] The flow sensor 10 observes the flow rate of the liquid food between the liquid transfer pump 3 and the heating heat exchanger 4. Temperature sensors 14 and 15 may be provided between the liquid transfer pump 3 and the heating heat exchanger 4, and between the holding tube 6 and the cooling heat exchanger 7 to observe the temperature of the liquid food (see Figure 2). Temperature sensor 11 observes the temperature of the liquid food between the heating heat exchanger 4 and the holding tube 6. Temperature sensor 12 observes the temperature of the liquid food between the cooling heat exchanger 7 and the back pressure regulator 8. Pressure sensor 13 observes the pressure of the liquid food between the cooling heat exchanger 7 and the back pressure regulator 8. The flow sensor 10, temperature sensor 11, temperature sensor 12, pressure sensor 13, and temperature sensors 14 and 15 transmit their observed values ​​to the control unit 20 of the control device 1.

[0028] The control device 1 comprises a control unit 20, an input device 21, and a target value setter 22. The input device 21 is a device for inputting and displaying the operating conditions of the continuous sterilization device 200. The input device 21 may be configured as a touch panel or the like. Examples of operating conditions include the type of product and differences in subsequent processes. Differences in product types include, for example, the sterilization temperature being different for coffee and tea. Differences in subsequent processes include differences in the amount to be filled, such as the flow rate being different for 2-liter PET bottles and 500cc PET bottles. The operating conditions input by the input device 21 are output to the target value setter 22. The target value setter 22 is a controller for outputting target values ​​for stable control. The target value setter 22 may be configured as a controller such as a general-purpose PLC. The target value setter 22 outputs target values ​​for stable control to the control unit 20. Target values ​​may be set for the flow rate of the liquid food by the liquid transfer pump 3, the temperature of the liquid food heated by the heating heat exchanger 4, the temperature of the liquid food cooled by the cooling heat exchanger 7, and the back pressure adjusted by the back pressure regulator 8.

[0029] The control unit 20 is a system that controls the operation of the entire continuous sterilization device 200. The control unit 20 is configured as a general-purpose computer, comprising a processor, memory, storage, a communication interface, and a user interface. The processor is an arithmetic unit such as a CPU (Central Processing Unit). The memory is a storage medium such as ROM (Read Only Memory) or RAM (Random Access Memory). The storage is a storage medium such as an HDD (Hard Disk Drive). The communication interface is a communication device that enables data communication. The user interface consists of output devices such as LCDs and speakers, and input devices such as control levers, buttons, keyboards, touch panels, and microphones. The processor integrates the memory, storage, communication interface, and user interface and realizes the functions described later. In the control unit 20, for example, various functions are realized by loading a program stored in ROM into RAM and executing the program loaded into RAM with the CPU. The control unit 20 may be composed of multiple computers.

[0030] The control unit 20 has the function of achieving stable control based on the target value for stable control set by the target value setter 22. The control unit 20 acquires parameters related to the temperature of the liquid or liquid food. Parameters related to the temperature of the liquid food include not only observed temperature values ​​but also parameters that affect the temperature. Specifically, parameters related to the temperature of the liquid food include observed values ​​of the flow rate of the liquid food and observed values ​​of back pressure.

[0031] The control unit 20 adjusts the control parameters of each control unit of the continuous sterilization device 200 based on the mutual influence of the acquired parameters. The control units are the devices that can be operated during the operation of the continuous sterilization device 200. In Figure 1, the liquid transfer pump 3, the heating heat exchanger 4, the cooling heat exchanger 7, and the back pressure regulator 8 are the control units. The control parameters for the control units are control parameters that indicate how much the control unit should be controlled. The control unit 20 calculates the control values ​​corresponding to each control unit as the control parameters for the control units. The mutual influence of acquired parameters refers to the influence that one parameter has on other parameters. If there are fluctuations in the acquired parameters, the control unit 20 performs coordinated control, which coordinates and integrates flow rate control, temperature control, and back pressure control to approach the target values ​​for stable control (see also Figure 2).

[0032] Here, the mutual influence of parameters and the adjustment of manipulated variables based on them will be explained with reference to Figure 3. Note that the contents shown in Figure 3 are merely an example and are not limited to this. For example, suppose the flow rate of liquid food increases. In this case, the expected secondary fluctuations are a decrease in the sterilization temperature (observed value from temperature sensor 11) and an increase in back pressure. In response to such parameter fluctuations, the control unit 20 performs coordinated control of control operations to decrease the flow rate, increase the sterilization temperature, and decrease the back pressure. Note that the flow rate of liquid food may increase or decrease. In this case, each control unit is controlled in the opposite direction.

[0033] For example, suppose the sterilization temperature of the liquid food decreases. In response to such parameter fluctuations, the control unit 20 performs coordinated control operations to decrease the flow rate, increase the sterilization temperature, and increase the back pressure. In this case, there are no expected secondary fluctuations, but by deliberately reducing the flow rate within the allowable range, the time the liquid food remains in the heating heat exchanger 4 increases, and the sterilization temperature can be temporarily raised in a shorter time than by operating the heating medium control valve 31 (see Figure 2). Note that the temperature of the liquid food may decrease or increase. In this case, each control unit is controlled in the opposite direction.

[0034] For example, suppose the back pressure of a liquid food increases. In this case, anticipated secondary fluctuations include an increase in sterilization temperature and a decrease in flow rate. In response to such parameter fluctuations, the control unit 20 performs coordinated control of increasing the flow rate, decreasing the sterilization temperature, and decreasing the back pressure. Note that while back pressure may increase, it may also decrease. In this case, each control unit is controlled in the opposite direction. Depending on the specifications of the device, back pressure fluctuations may not affect the flow rate and sterilization temperature. In this case, coordinated control of flow rate and sterilization temperature is not necessary.

[0035] The control unit 20 modifies the control values ​​(operated quantities) for each control unit while acquiring parameters related to the temperature of the liquid food during the operation of the continuous sterilization device 200. Furthermore, by acquiring parameters, the control unit 20 predicts the state of the continuous sterilization device 200 after a finite length of time has elapsed, and modifies the controlled quantities based on this prediction. The control unit 20 can predict parameters for a short period of time ahead. Based on this prediction, the control unit 20 modifies the controlled quantities. The control unit 20 controls the temperature of the liquid food in the continuous sterilization device 200 using a prediction model. The prediction model used by the control unit 20 may be a prediction model automatically learned from a database storing past controlled values ​​and observed values ​​using machine learning techniques such as neural networks. The control unit 20 can use such a prediction model by incorporating artificial intelligence.

[0036] Next, the detailed configuration of the control unit 20 will be described with reference to Figure 4. Figure 4 is a diagram showing the block configuration of the control unit 20. As shown in Figure 4, the control unit 20 comprises a learning unit 41, a storage unit 42, an optimization unit 43, a prediction model unit 44, a prediction error correction unit 46, and a calculation unit 47.

[0037] The learning unit 41 is a unit that learns prediction model parameters. The learning unit 41 retrieves learning data from the database stored in the memory unit 42 and performs learning. The learning unit 41 performs learning using methods such as backpropagation. The learning unit 41 outputs the prediction model parameters obtained based on the learning to the prediction model unit 44. As a result, the prediction model unit 44 is improved (modified) to an optimal state.

[0038] The memory unit 42 is a unit that stores various types of information. The memory unit 42 stores past operational values ​​and observed values. The memory unit 42 builds a database by accumulating operational values ​​and observed values ​​for several days to several years. The memory unit 42 outputs learning data to the learning unit 41 for the learning unit 41 to learn from.

[0039] The optimization unit 43 is a unit that optimizes the control values ​​for the operation unit and determines the control values. The optimization unit 43 determines the control values ​​using methods such as sequential quadratic programming. The optimization unit 43 determines provisional control values. By repeating the optimization process, the optimization unit 43 can bring the control values ​​closer to the optimal values. The optimization unit 43 outputs the provisional control values ​​to the prediction model unit 44.

[0040] The prediction model unit 44 is a unit that acquires predicted values ​​using a prediction model. The prediction model unit 44 acquires current observed values ​​such as flow rate, temperature, and back pressure from each sensor and uses a neural network or the like to acquire predicted values ​​of flow rate, temperature, and back pressure after a finite length of time has elapsed. The prediction model unit 44 acquires the parameters obtained when provisional operating values ​​acquired from the optimization unit 43 are applied to the prediction model as predicted values. Once the prediction model unit 44 has acquired the predicted values, it outputs them to the prediction error correction unit 46.

[0041] The prediction error correction unit 46 is a unit that corrects the error in the predicted value set by the prediction model unit 44. Prediction models using machine learning methods are subject to some error due to factors such as the degree of learning. Therefore, the prediction error correction unit 46 corrects the error in the predicted value using an error corrector such as a Kalman filter. The prediction error correction unit 46 outputs the error-corrected predicted value to the calculation unit 47.

[0042] The calculation unit 47 performs various calculations in the control unit 20. The calculation unit 47 obtains the target setting value from the input device 21 or the target value setter 22. The calculation unit 47 calculates the difference between the predicted value and the target value and determines whether the difference is minimized. If the calculation unit 47 determines that the difference is not minimized, it sends an optimization command to the optimization unit 43 to obtain an operation value that can reduce the difference. If the calculation unit 47 determines that the difference is minimized, it outputs the currently set operation value to the target.

[0043] Each control unit (each pump, valve, etc.) operates based on the control value obtained from the calculation unit 47. As a result, the control units are operated according to the control value. The control units provide control outputs for control units such as temperature and flow rate. Each sensor, such as the temperature sensor and flow rate sensor, outputs its observed values ​​to the storage unit 42, the prediction model unit 44, and the prediction error correction unit 46.

[0044] Even if there are two or more targets to be operated on, they can be operated with a single control device 1. However, two or more control devices 1 may be used to distribute the load. The operated value, provisional operated value, observed value, and predicted value are finite-length time-series data, and their length may be adjusted according to the specifications of the device.

[0045] Next, the processing details of the control unit 20 will be explained with reference to Figures 5 and 6. Figure 5 is a flowchart showing the learning process. Figure 6 is a flowchart showing the processing details of the control unit 20.

[0046] The trigger for starting the process shown in Figure 5 may be manual or automatically started by pre-setting the number of times per day. Furthermore, the process shown in Figure 5 may be performed in parallel with the real-time operation shown in Figure 6. As shown in Figure 5, the learning unit 41 reads the learning data for the learning period from the database (step S10). The period for which the learning data is extracted may be adjusted according to the specifications of the device. Next, the learning unit 41 learns the prediction model parameters (step S20). Next, the learning unit 41 evaluates the prediction model using the learned parameters, the most recent observed values ​​from the database, and the manipulated values ​​(step S30). The learning unit 41 determines whether the evaluated prediction model is superior to the current prediction model (step S40). If it is determined to be superior, the learning unit 41 updates and saves the prediction model parameters used in the real-time operation (step S50) and terminates the process shown in Figure 5. If it is determined to be inferior, the process shown in Figure 5 terminates without performing the process in step S50.

[0047] Next, the control process by the control unit 20 will be described with reference to Figure 6. This process is performed in real time during the sterilization operation of the continuous sterilization device 200. First, the prediction model unit 44 determines whether or not newly learned prediction model parameters exist (step S100). If it determines that they exist, the prediction model unit 44 updates the prediction model parameters (step S110). If it determines that they do not exist, the prediction model unit 44 proceeds to the next step without updating. Next, the prediction model unit 44 initializes the provisional operating values ​​(step S120).

[0048] Next, the control unit 20 acquires the observed values ​​from each sensor and the target value (step S130). Next, the prediction model unit 44 calculates a predicted value from the observed values ​​and the provisional operating value (step S140). Next, the prediction error correction unit 46 corrects the error in the predicted value predicted in step S140 from the predicted value and the observed value (step S150). Next, the calculation unit 47 calculates the difference between the predicted value corrected in step S150 and the target value (step S170).

[0049] If it is determined in step S170 that the difference is not minimized, the calculation unit 47 outputs an optimization command to the optimization unit 43, and the optimization unit 43 updates the temporary operation values ​​so that the difference is further reduced (step S180). Once step S180 is executed, the process is repeated again from step S130. If it is determined in step S170 that the difference is minimized, the calculation unit 47 outputs the temporary operation values ​​as operation values ​​to each operation unit (step S190). The calculation unit 47 also outputs the operation values ​​to the storage unit 42, and the storage unit 42 saves the operation values ​​and observed values ​​linked together in the database (step S200). Next, the control unit 20 waits for a certain period of time to elapse based on a preset waiting time (step S210). Once the period of time has elapsed, the control unit 20 repeats the process again from step S100.

[0050] Next, the operation and effects of the control device 1 and the food processing machine 100 according to this embodiment will be described.

[0051] First, with reference to Figure 7, the comparative example food processing machine 300 will be described. Instead of the control unit 20 described above, the food processing machine 300 is equipped with a controller 51 for flow rate control, controllers 52 and 54 for temperature control, and a controller 53 for back pressure control. Each controller 51, 52, 53, and 54 performs PID control. Here, each controller 51, 52, 53, and 54 acquires observed values ​​from sensors for the parameters of the controlled object and outputs the operated values ​​to the control object's operating unit. The controllers 51, 52, 53, and 54 control the controlled object individually and do not consider other parameters. Therefore, in the comparative example food processing machine 300, temperature control by PID control cannot take into account disturbances such as flow rate fluctuations and back pressure fluctuations, which sometimes resulted in a decrease in the accuracy of temperature control of the heating heat exchanger 4.

[0052] In the comparative example food processing machine 300, maintenance causes a change in the state of the food processing machine 300. To address this change in state, it is necessary to adjust the control parameters. However, adjusting such parameters relies on the experience and intuition of the operator, which is difficult for the average operator. As a result, there was a problem of reduced accuracy in the temperature control of the control device. In addition, PID control requires a certain amount of time to stabilize and converge, which results in wasted time and energy being consumed during that period.

[0053] In contrast, the control device 1 according to this embodiment acquires parameters related to the liquid temperature and adjusts the operating amount of each control unit of the food processing machine 100 based on the mutual influence of these parameters. In this case, the control device 1 can adjust the operating amount of each control unit while considering the influence of other temperature-related parameters (such as flow rate and back pressure) in addition to the liquid temperature. Therefore, the control device 1 can control each control unit more appropriately than when only the liquid temperature is considered. As a result, the accuracy of liquid temperature control can be improved.

[0054] The control device 1 may modify the manipulated variables while acquiring parameters during the operation of the food processing machine 100. In this case, the control device 1 can perform accurate temperature control in real time during the operation of the food processing machine 100.

[0055] The control device 1 may predict the state of the food processing machine 100 after a finite length of time has elapsed by acquiring parameters, and may modify the manipulated variable based on the prediction result. In this case, the control device 1 can anticipate the state of the food processing machine 100 and take preventative measures before the liquid deviates from the appropriate temperature.

[0056] The control device 1 may control the temperature of the liquid in the food processing machine 100 using a predictive model. In this case, the control device 1 can easily and accurately control the temperature by applying the observed values ​​of acquired parameters and the manipulated quantities for each control unit to the predictive model.

[0057] You may train the predictive model at least once a day. This will help keep the predictive model in an optimal state.

[0058] The liquid food processing machine 100 is equipped with the control device 1 described above.

[0059] The control method is a control method for controlling the temperature of a liquid or liquid food in a food processing machine for liquid foods, and involves acquiring parameters related to the temperature of the liquid and adjusting the operating amount of each operating part of the food processing machine 100 based on the mutual influence of the parameters.

[0060] According to the food processing machine 100 and control method, the same effects and benefits as those of the control device 1 described above can be obtained.

[0061] Furthermore, according to the control device 1 described above, by using machine learning techniques, the adjustment of control parameters can be automated, and adjustments to respond to changes in the state of the food processing machine 100 can be made automatically with minimal or no human intervention. In addition, unlike PID control, the control device 1 can optimize the time required for stable convergence and the operating values ​​by calculating appropriate operating values ​​from the prediction results. Therefore, time reduction and energy saving effects can be obtained. In particular, in the continuous sterilization device 200, fluctuations in sterilization temperature can cause insufficient sterilization due to temperature drops or quality deterioration due to temperature increases, so the permissible range of temperature fluctuations is very narrow. Therefore, the effect of stable control of the sterilization temperature by the control device 1 is even more significant.

[0062] The present invention is not limited to the embodiments described above.

[0063] While a continuous sterilization device was given as an example of food processing machinery, the present invention may be applied to any food processing machinery related to liquid foods. For example, it may be applied to a wide range of food processing machinery for liquid foods, such as CIP devices, hot water heating devices, and cooling devices.

[0064] Figure 8 is a system diagram showing the case where the CIP device 400 is used as the food processing machine 100. As shown in Figure 8, the CIP device 400 has a configuration similar in purpose to the configuration upstream of the holding tube 6 of the continuous sterilization device 200 shown in Figure 2. However, the CIP device 400 has a cleaning solution tank 62 for storing cleaning solution instead of the balance tank 2. The control unit 20 performs coordinated control of flow rate control and temperature control. The cleaning solution heated in the CIP device 400 flows through the supply line L1, and the supply destination is switched at the switching unit 63 to either the food production line 150A or the food production line 150B. The cleaning solution is supplied to one of the items to be cleaned and cleaning is performed. The cleaning solution is returned to the cleaning solution tank 62 via line L2. However, the configuration in Figure 8 is an example and may be changed as appropriate depending on the specifications.

[0065] The CIP device 400 described above is a system that cleans the inside of equipment without disassembling the food manufacturing line, while the equipment remains in the state it was in during manufacturing. This CIP device 400 is a cleaning device installed separately from the food manufacturing line, such as a continuous sterilization device. Due to the following features, the CIP device 400 achieves efficient cleaning and improves the safety and quality of food manufacturing. Specifically, the CIP device 400 has an automated cleaning process. The CIP device 400 can automatically manage the supply, flow rate, temperature, and time of the cleaning solution using the control device 1 according to this embodiment. The CIP device 400 has a multi-stage cleaning process. The CIP device 400 has multiple processes such as cleaning, rinsing, and sterilization, and can use different cleaning solutions in each process or change to the optimal temperature conditions for each process. The CIP device 400 can perform closed-type cleaning. The CIP device 400 can clean the inside of the equipment while preventing contamination from the external environment. The CIP device 400 can use reusable cleaning solution. In some cases, the cleaning solution can be reused in the CIP device 400 to save costs and be environmentally friendly.

[0066] In conventional feedback control for stabilizing the temperature and flow rate of the cleaning fluid supplied by a CIP (Clean-in-Place) system, PID control was used to calculate operating values ​​such as the steam valve and pump speed of the heat transfer medium. PID control, similar to that used in the continuous sterilization system explained in Figure 7, has the problem of being susceptible to external disturbances such as flow rate, limiting its ability to suppress temperature fluctuations. Furthermore, adjustments to respond to changes in the system's condition due to system maintenance are made by adjusting the PID parameters, but adjusting the PID control parameters relies on the operator's experience and intuition, making it difficult for operators to perform the adjustments. In addition, in conventional CIP systems, a decrease in the cleaning fluid temperature can impair the cleaning effect, and a decrease in the cleaning fluid flow rate can lead to insufficient cleaning. When temperature or flow rate drops, wasted time is created for stabilization and convergence, which can extend the cleaning time, thus requiring stable control of the cleaning fluid supply.

[0067] In contrast, the CIP device 400 shown in Figure 8 incorporates artificial intelligence, allowing it to predict the temperature of the cleaning fluid and open and close the steam valve before temperature fluctuations occur. Furthermore, by cooperatively controlling temperature and flow rate, the CIP device 400 can minimize the effects of disturbance fluctuations, leading to improved stability of the cleaning fluid temperature and supply flow rate. While PID control requires some time for stable convergence, the CIP device 400 can optimize the time and amount of operation required for stable convergence by calculating the optimal operation amount from the predictive results, resulting in reduced time and energy savings.

[0068] The CIP device 400 automates the adjustment of control parameters using machine learning techniques, enabling it to automatically adjust to changes in the device's condition due to maintenance or other reasons with minimal or no human intervention. Furthermore, the CIP device 400 learns from information on the switching of cleaning routes in the food production line being cleaned, allowing it to supply the optimal liquid for each cleaning route. These advantages are particularly pronounced in the CIP device 400.

[0069] Note that the continuous sterilization device 200 shown in Figure 2 may also perform cleaning using the heat exchanger 4 originally installed in the device, without using the CIP device 400. In this case, the continuous sterilization device 200 stores cleaning solution in the balance tank 2 instead of liquid food and flows it downstream. The control valve of the back pressure regulator 8 is left fully open, and the control unit 20 does not control the back pressure. The cleaning solution is returned from the back pressure regulator 8 to the balance tank and circulated. The control unit 20 performs coordinated control of flow rate and temperature to maintain the cleaning solution at a stable cleaning temperature.

[0070] When cleaning the heat exchanger originally installed in a continuous sterilization device without using a CIP device, conventional methods used PID control, which resulted in the same problems as with a CIP device. However, by adopting the control unit 20, the same advantages as a CIP device can be obtained. Furthermore, the continuous sterilization device 200 can perform optimal control for each mode by learning from the mode switching information during product production and cleaning.

[0071] As shown in Figure 9, a mechanism may be employed to return liquid food to the balance tank 2. The continuous sterilization device 200 shown in Figure 9 has a switching valve 71 that switches the flow path between the back pressure regulator 8 and the cooling heat exchanger 7. A return line L3 extends from the switching valve 71 to the balance tank 2. The continuous sterilization device 200 also has a switching valve 72 between the holding tube 6 and the cooling heat exchanger 7. A return line L4 extends from the switching valve 72 and merges with the return line L3. A control valve 73 is provided on the balance tank side of the merging point of return lines L3 and L4. The continuous sterilization device 200 has a switching unit 74 that switches the output destination of the operation value related to back pressure control of the control unit 20 between the control valve of the back pressure regulator 8 and the control valve 73 on the return line L3.

[0072] For example, in Figure 9, if the sterilization temperature decreases or the liquid flow rate increases, the control unit 20 switches the switching valve 72 to return the liquid food to the balance tank 2 via the return lines L4 and L3. If a malfunction occurs in the next process or later and the next process tank 34 becomes full, the control unit 20 switches the switching valve 71 to return the liquid food to the balance tank 2 via the return line L3. The control unit 20 switches the switching unit 74 according to the status of the switching valves 71 and 72 to switch the control valve to be controlled. The control unit 20 controls the control valve to an optimal opening degree that minimizes fluctuations in back pressure and flow rate.

[0073] Furthermore, the control unit 20 may prepare multiple prediction models. The control unit 20 may switch between prediction models based on the conditions in the food processing machine 100. In this case, the control device 1 can perform temperature control with high accuracy according to the conditions of the food processing machine 100. For example, if the type of liquid food is changed, the control unit 20 may prepare a prediction model corresponding to the liquid food.

[0074] For example, multiple learning units 41 and multiple storage units 42 as shown in Figure 10 may be applied to the learning unit 41 and storage unit 42 shown in Figure 4, and the prediction model may be switched depending on the conditions. For example, the prediction model may be switched for each type of liquid food. Specifically, when manufacturing liquid food A, the control device 1 uses database A of storage unit 42A and learning unit 41A; when manufacturing liquid food B, it uses database B of storage unit 42B and learning unit 41B; and when manufacturing liquid food C, it uses database C of storage unit 42C and learning unit 41C. The control device 1 includes a switching unit 66 that switches the storage unit 42 to which the observed values ​​are output, a switching unit 67 that switches the storage unit 42 to which the operated values ​​are output, and a switching unit 68 that switches the learning unit 41 that is the source of the selected prediction model parameters, depending on the prediction model used. Since parameter settings such as sterilization temperature and flow rate, and physical property information such as viscosity and specific heat differ for each type of liquid food, the control device 1 can switch databases to achieve optimal control for each condition.

[0075] The control device 1 may switch the prediction model depending on the item to be cleaned when switching the item to be cleaned in the CIP device 400. The control device 1 may also switch the prediction model depending on whether the continuous sterilization device 200 is flowing liquid food or cleaning solution. Furthermore, as shown in Figure 9, the control device 1 may switch the prediction model between normal operation and when returning liquid food. In addition, the prediction model may be switched when switching conditions such as the type of cleaning solution (detergent) when flowing cleaning solution as a condition of the food processing machine 100. Furthermore, the prediction model can be switched depending on the type of product during normal operation of the sterilization device.

[0076] The control device 1 may retrain its predictive model during maintenance of the food processing machine 100 or during line modifications. In this case, the control device 1 can perform temperature control with high accuracy, taking into account the effects of maintenance or line modifications.

[0077] For example, product flow rate and pressure may change compared to normal (when liquid transfer pump 3 is functioning normally and operating without replacement or maintenance) due to replacement or maintenance of the liquid transfer pump 3. In response to this, the control device 1 may be optimized by relearning the predictive model. Due to replacement or maintenance of pump 33, the flow rate and pressure of the heat transfer medium may change compared to normal (when pump 33 is functioning normally and operating without replacement or maintenance), which may change the steam inflow rate and heat exchange conditions, affecting the product temperature. In response to this, the control device 1 may be optimized by relearning the predictive model. Due to replacement or maintenance of the control valve 31, the steam flow rate and pressure may change compared to normal (when control valve 31 is functioning normally and operating without replacement or maintenance). In response to this, the control device 1 may be optimized by relearning the predictive model. Due to extension or rerouting of the holding tube 6 or other piping, product flow rate and pressure may change compared to normal (when the holding tube 6 or other piping are operating without replacement or maintenance). In response to this, the control device 1 may be optimized by relearning the predictive model. Due to replacement or maintenance of the control valve of the back pressure regulator 8, the product flow rate and pressure may change compared to normal (a state in which the control valve is functioning normally and is operating without replacement or maintenance). In response to this, the control device 1 may be optimized by relearning the predictive model. [Explanation of symbols]

[0078] 1...Control device, 100...Food processing machinery, 200...Continuous sterilization device, 400...CIP device.

Claims

1. A control device for controlling the temperature of a liquid or liquid food in a food processing machine for liquid foods, Obtain parameters related to the temperature of the aforementioned liquid or liquid food, A control device that adjusts the amount of operation of each operating part of the food processing machine based on the mutual influence of the aforementioned parameters.

2. The control device according to claim 1, which modifies the manipulated amount while acquiring the parameters during the operation of the food processing machine.

3. The control device according to claim 1, which predicts the state of the food processing machine after a finite length of time has elapsed by acquiring the aforementioned parameters, and modifies the manipulated variable based on the prediction result.

4. The control device according to claim 1, which controls the temperature of the liquid or liquid food in the food processing machine using a predictive model.

5. The control device according to claim 4, which periodically performs training of the prediction model at least once a day.

6. Prepare multiple predictive models, The control device according to claim 4, which switches the prediction model based on the conditions in the food processing machine.

7. The control device according to claim 4, wherein the predictive model is retrained when the food processing machine is being maintained or the production line is being modified.

8. A food processing machine for liquid food, comprising a control device as described in any one of claims 1 to 7.

9. A control method for controlling the temperature of a liquid or liquid food in a food processing machine for liquid foods, Obtain parameters related to the temperature of the aforementioned liquid or liquid food, A control method for adjusting the amount of operation of each operating part of the food processing machine based on the mutual influence of the aforementioned parameters.

Images