Intelligent regulation and control system and method for meal supply quantity of intelligent canteen

Abstract

The invention provides an intelligent regulation and control system and method for the meal supply amount of an intelligent canteen. The system comprises a multi-modal sensing unit, a computing device and an actuator, the computing device comprises a coupling control unit and a control module group, and the control module group comprises a compensator, a suppressor and a trigger which are respectively used for performing real-time compensation on an integral time constant of a proportional-integral control algorithm according to a dynamic correlation weight matrix, calculating the required meal supplement amount; calculating a feedforward compensation amount in a future time window based on the inventory weight change data and the dynamic consumption rate fluctuation data in the sliding time window; when the real-time consumption rate exceeds a preset consumption rate threshold value, generating a pulse width modulation signal of which the duty ratio is positively correlated with the consumption rate gradient; and the actuator executes meal supplement operation according to the meal supplement amount, the feed-forward compensation amount and the pulse width modulation signal so as to form a closed-loop control loop. According to the invention, the accuracy of intelligent regulation and control of the meal supply quantity can be realized.

Description

Technical Field

[0001] This application relates to the technical field of catering management, and particularly to an intelligent regulation system and method for the supply quantity of food in a smart cafeteria. Background Art

[0002] In the scenario of a smart cafeteria, the precise regulation of the supply quantity of food is the core requirement for improving dining efficiency and reducing waste. Specifically, this scenario needs to meet the dynamic response ability, collaborative consumption modeling, and adaptive control.

[0003] To address the above requirements, the prior art adopts the following three solutions. The first solution is single-sensor threshold triggering. A weighing sensor is deployed at the bottom of the food container, and when the inventory weight is lower than the preset threshold, a replenishment signal is triggered, which can achieve basic inventory warning and replace manual inspection. The second solution is regular meal replenishment statistics. A meal replenishment plan is formulated based on historical consumption data, and replenishment is executed according to a fixed time window (such as every 2 hours). The third solution is fixed parameter proportional-integral-derivative (PID) control.

[0004] However, the prior art has problems of untimely supply response or over-replenishment. Summary of the Invention

[0005] Embodiments of this application provide an intelligent regulation system and method for the supply quantity of food in a smart cafeteria, which are used to solve the problem of poor accuracy of intelligent regulation in the prior art.

[0006] In a first aspect, embodiments of this application provide an intelligent regulation system for the supply quantity of food in a smart cafeteria, including: A multimodal sensing unit, including a weighing sensor, a plate radio frequency reader, and a visual recognition device arranged at the bottom of the food container, which is used to collect real-time data on the change in the inventory weight of food, food identification data, and dynamic consumption correlation data, and the dynamic consumption correlation data represents the collaborative consumption time sequence relationship between foods; A computing device includes a coupled control unit and a control module group. The coupled control unit is used to establish a dynamic association weight matrix through the real-time data stream of the multimodal sensing unit, and calculate the real-time consumption rate based on the gradient feedback signal of the change in inventory weight and the deviation of the replenishment quantity of the actuator. The dynamic association weight matrix is used to represent the collaborative consumption characteristics between foods within a closed-loop control period; The control module group includes a compensator, a suppressor, and a trigger; The compensator is used to perform real-time compensation on the integral time constant of the proportional-integral (PI) control algorithm according to the dynamic association weight matrix, and calculate the required replenishment quantity; The suppressor is used to calculate the feedforward compensation amount within a future time window based on the inventory weight change data and dynamic consumption rate fluctuation data within a sliding time window. The trigger is used to generate a Pulse Width Modulation (PWM) signal with a duty cycle positively correlated with the consumption rate gradient when the real-time consumption rate exceeds a preset consumption rate threshold. The actuator is used to drive the meal replenishment robotic arm within the actuator to perform a meal replenishment operation according to the meal replenishment amount, feedforward compensation amount, and Pulse Width Modulation signal through an industrial bus protocol, and feedback and update the corresponding inventory data through the multimodal sensing unit to form a closed-loop control loop.

[0007] Optionally, the coupled control unit includes a weight matrix generation module and a real-time consumption rate prediction module connected in sequence. The coupled control unit is used to establish a dynamic association weight matrix through the real-time data stream of the multimodal sensing unit, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the meal replenishment amount deviation of the actuator. The dynamic association weight matrix is used to characterize the collaborative consumption characteristics between food items within a closed-loop control cycle, including: The weight matrix generation module is used to generate a dynamic weight matrix through dynamic coupling relationship analysis within a sliding time window based on the real-time collected food item identification data and dynamic consumption association data, where the update period of the dynamic weight matrix is consistent with the sampling period of the proportional-integral control algorithm. The real-time consumption rate prediction module is used to calculate the instantaneous consumption rate within the current time window according to the gradient change of the inventory weight change data, and perform a convolution operation on the instantaneous consumption rate and the dynamic weight matrix based on the error transfer function of the proportional-integral controller to obtain the real-time consumption rate.

[0008] Optionally, the weight matrix generation module includes a data processing sub-module, a timestamp recording sub-module, a hardware accelerator, and a matrix operation sub-module. The weight matrix generation module is used to generate a dynamic weight matrix through dynamic coupling relationship analysis within a sliding time window based on the real-time collected food item identification data and dynamic consumption association data, including: The data processing sub-module is used to group and cache the food item identification data within the same meal selection combination based on the length of the time window, and generate a dynamic consumption association event data packet. The timestamp recording sub-module is used to add a corresponding timestamp to each dynamic consumption association event data packet, and trigger an incremental update of the coupling coefficient when the timestamp interval is less than the preset control period. The hardware accelerator is used to perform real-time encoding on the meal identification data through a parallel computing architecture to generate a meal feature vector, and dynamically update the combined feature vector based on the sampling period of the multimodal sensing unit; The matrix operation sub-module is used to perform a tensor product operation on the combined feature vector and the historical consumption rate data to generate a dynamic weight matrix that conforms to the industrial fieldbus data frame structure.

[0009] Optionally, the real-time consumption rate prediction module is used to calculate the instantaneous consumption rate within the current time window according to the gradient change of the inventory weight change data, and perform a convolution operation on the instantaneous consumption rate and the dynamic weight matrix based on the error transfer function of the proportional-integral controller to obtain the real-time consumption rate, including: Design a convolution kernel function based on the error transfer function of the proportional-integral controller; Use the convolution kernel function to perform a discrete convolution operation on the instantaneous consumption rate and the dynamic weight matrix to obtain the real-time consumption rate, and the real-time consumption rate and the replenishment meal quantity deviation jointly adjust the proportional gain and integral gain of the error transfer function.

[0010] Optionally, the matrix operation sub-module is used to perform a tensor product operation on the combined feature vector and the historical consumption rate data to generate a dynamic weight matrix that conforms to the industrial fieldbus data frame structure, including: The matrix operation sub-module is used to perform a tensor product operation on the combined feature vector and the historical consumption rate data to generate an initial weight matrix; The matrix operation sub-module is further used to perform normalization processing on the initial weight matrix, and encapsulate the normalized weight matrix according to the data frame format of the industrial fieldbus protocol to obtain the dynamic weight matrix.

[0011] Optionally, the actuator further includes a task scheduling module; The actuator is used to perform a replenishment meal operation according to the replenishment meal quantity, the feedforward compensation quantity, and the pulse width modulation signal, including: The task scheduling module is used to perform multi-task scheduling on the replenishment meal quantity, the feedforward compensation quantity, and the pulse width modulation signal to obtain a task scheduling result; The replenishment meal robotic arm is used to perform a corresponding replenishment meal operation according to the task scheduling result.

[0012] Optionally, the task scheduling module is used to perform multi-task scheduling on the replenishment meal quantity, the feedforward compensation quantity, and the pulse width modulation signal to obtain a task scheduling result, including: The task scheduling module is used to assign corresponding priority coefficients to the replenishment meal quantity, the feedforward compensation quantity, and the pulse width modulation signal respectively; The task scheduling module is further configured to sort according to the priority coefficients in descending order to generate a task scheduling result.

[0013] In a second aspect, an embodiment of the present application provides an intelligent regulation method for the supply quantity of food items in a smart cafeteria, including: Obtaining the real-time data stream of the multi-modal sensing unit through the coupling control unit, establishing a dynamic correlation weight matrix, and calculating the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the deviation of the replenishment quantity of the actuator. The dynamic correlation weight matrix is used to characterize the collaborative consumption characteristics between food items within a closed-loop control period; Using the compensator to perform real-time compensation on the integral time constant of the proportional-integral control algorithm according to the dynamic correlation weight matrix, and calculating the required replenishment quantity; Using the suppressor to calculate the feedforward compensation quantity within a future time window based on the inventory weight change data and the dynamic consumption rate fluctuation data within a sliding time window; Using the trigger to generate a pulse width modulation signal whose duty cycle is positively correlated with the consumption rate gradient when the real-time consumption rate exceeds a preset consumption rate threshold; Sending the replenishment quantity, the feedforward compensation quantity, and the pulse width modulation signal to the actuator, so that the actuator drives the replenishment robotic arm in the actuator to perform a replenishment operation according to the replenishment quantity, the feedforward compensation quantity, and the pulse width modulation signal through an industrial bus protocol, and feeds back and updates the corresponding inventory data through the multi-modal sensing unit to form a closed-loop control loop.

[0014] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an intelligent regulation method for the supply quantity of food items in a smart cafeteria as described in any item of the second aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it implements an intelligent regulation method for the supply quantity of food items in a smart cafeteria as described in any item of the second aspect.

[0016] In an embodiment of the present application, an intelligent regulation system for the supply quantity of food items in a smart cafeteria is provided, and the system includes: The multimodal sensing unit includes a weighing sensor, a plate radio frequency reader and writer, and a visual recognition device arranged at the bottom of the food container, and is used to collect real-time inventory weight change data, food identification data and dynamic consumption association data of the food. The dynamic consumption association data characterizes the collaborative consumption timing relationship between foods; the computing device includes a coupling control unit and a control module group. The coupling control unit is used to establish a dynamic association weight matrix through the real-time data stream of the multimodal sensing unit, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the replenishment quantity deviation of the actuator. The dynamic association weight matrix is used to characterize the collaborative consumption characteristics between foods within the closed-loop control cycle; the control module group includes a compensator, a suppressor and a trigger; the compensator is used to compensate the integral time constant of the proportional-integral control algorithm in real time according to the dynamic association weight matrix, and calculate the required replenishment quantity; the suppressor is used to calculate the feedforward compensation quantity within the future time window based on the inventory weight change data and the dynamic consumption rate fluctuation data within the sliding time window; the trigger is used to generate a pulse width modulation signal whose duty cycle is positively correlated with the consumption rate gradient when the real-time consumption rate exceeds the preset consumption rate threshold; the actuator is used to drive the replenishment robotic arm in the actuator to perform a replenishment operation according to the replenishment quantity, the feedforward compensation quantity and the pulse width modulation signal, and feedback and update the corresponding inventory data through the multimodal sensing unit to form a closed-loop control loop.

[0017] In the embodiment of the present application, through the three-dimensional data fusion of the weighing sensor, the radio frequency reader and writer, and the visual recognition, the dynamic capture of the food inventory quantity, identification information and consumption timing is realized, the limitation of the single sensing mode is broken through, the spatio-temporal association mapping between the weight change gradient and the consumption behavior is established, and the collaborative consumption characteristics of foods are modeled through the dynamic association weight matrix, providing a multi-dimensional feature space for intelligent replenishment. Based on the time-varying parameter compensation mechanism of proportional-integral control, the dynamic matching of the control algorithm parameters and the consumption rate is realized. The sliding time window mechanism combined with the calculation of the feedforward compensation quantity effectively responds to the non-steady-state fluctuation of the consumption demand. The generation of the pulse width modulation signal realizes the gradient response of the replenishment intensity, forming a collaborative control of multiple time scales. A "compensation-suppression-trigger" trinity control architecture is constructed to meet the dual requirements of steady-state regulation and transient response at the same time. Through the actuator driven by the industrial bus protocol, a millisecond-level closed-loop response from the replenishment decision to the robotic arm action is realized. The continuous iterative update of the dynamic consumption association data forms an intelligent inventory management system with self-learning ability.

[0018] Furthermore, a dynamic association weight matrix is constructed in real time through multi-modal sensing data streams. Based on a sliding time window, the dynamic coupling relationship between meal identification data and consumption association events is analyzed. A hardware accelerator is used to encode and update feature vectors, and a dynamic weight matrix conforming to the industrial bus protocol is generated through tensor product operations. At the same time, combining the gradient feedback of inventory weight changes with the error transfer function of a proportional-integral controller, convolution operations are used to fuse the instantaneous consumption rate and the weight matrix, realizing the modeling of the collaborative consumption characteristics of meals within a closed-loop cycle and the prediction of the real-time consumption rate. The matrix update period is strictly synchronized with the control system sampling. The collaborative consumption timing relationship between meals is accurately characterized by the dynamic weight matrix, and the feature processing delay is compressed to the millisecond level using a hardware-accelerated parallel computing architecture, enabling the system to dynamically perceive the associated impact of combined meal selection behaviors on inventory. Combining with a convolution prediction algorithm based on controller error transfer reduces the consumption rate prediction error. At the same time, through the matrix data frame standardized by the industrial bus protocol, high-reliability real-time interaction between control instructions and actuators is achieved, ultimately forming an adaptive closed-loop response ability to non-linear demand fluctuations in complex catering consumption scenarios, improving the accuracy of meal replenishment timing matching compared to traditional methods.

[0019] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] To more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 It is a schematic structural diagram of an intelligent regulation system for the supply quantity of meals in a smart cafeteria provided by an embodiment of the present application; Figure 2 It is a flowchart of an intelligent regulation method for the supply quantity of meals in a smart cafeteria provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to enable those skilled in the art to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0023] In some processes described in the specification, claims, and abovementioned drawings of this application, multiple operations appear in a specific order. However, it should be clearly understood that these operations can be executed not in the order in which they appear herein or in parallel. The operation numbers such as 11, 12, etc. are only used to distinguish different operations, and the numbers themselves do not represent any execution order. Additionally, these processes can include more or fewer operations, and these operations can be executed sequentially or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0024] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope protected by this application.

[0025] Figure 1 It is a schematic structural diagram of an intelligent regulation system for the supply quantity of food in a smart cafeteria provided by an embodiment of this application. As Figure 1 shown, the system includes: A multimodal sensing unit 10, including a weighing sensor, a tray radio frequency reader / writer, and a visual recognition device arranged at the bottom of the food container, is used to collect real-time data on the change in the inventory weight of food, food identification data, and dynamic consumption correlation data. The dynamic consumption correlation data represents the collaborative consumption timing relationship between foods.

[0026] Among them, the weighing sensor converts the weight change into an electrical signal, converts it into weight data through existing algorithms, and calculates the consumption amount through the weight difference. For example, the initial weight is 1 kg, and it decreases to 0.8 kg, indicating a consumption of 200 g. The tray radio frequency reader / writer reads / writes the label information on the tray through radio frequency signals. For example, such as food identification codes, categories, etc. The visual recognition device distinguishes different foods through features such as shape and color. Further, the visual recognition device can be an identification model including a camera and YOLOv5, and is used to obtain food identification data and dynamic consumption correlation data.

[0027] The computing device 11 includes a coupling control unit and a control module group. The coupling control unit is connected to the multimodal sensing unit through an industrial fieldbus, and is used to establish a dynamic association weight matrix through the real-time data stream of the multimodal sensing unit, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the replenishment amount deviation of the actuator. The dynamic association weight matrix is used to represent the collaborative consumption characteristics between foods within the closed-loop control cycle.

[0028] A control module group, including a compensator, a suppressor and a trigger; the compensator is used to compensate the integral time constant of the proportional-integral control algorithm in real time according to the dynamic correlation weight matrix and calculate the required replenishment amount; the suppressor is used to calculate the feedforward compensation amount in the future time window based on the inventory weight change data and the dynamic consumption rate fluctuation data within the sliding time window; the trigger is used to generate a pulse width modulation signal whose duty cycle is positively correlated with the consumption rate gradient when the real-time consumption rate exceeds the preset consumption rate threshold. The compensator adjusts the control algorithm parameters through the dynamic weight matrix to cope with the current fluctuations, and the suppressor predicts future demands based on historical data for early compensation. The two cooperate and optimize for real-time error and lag error respectively. "The duty cycle is positively correlated with the consumption rate gradient" can be understood as: the duty cycle of the pulse width modulation signal is equal to the real-time consumption rate gradient multiplied by a preset multiple, and the preset multiple can be determined according to the actual situation, such as 2, 3, etc.

[0029] Among them, the dynamic correlation weight matrix is a matrix that quantifies the collaborative consumption intensity between food items, and the larger the element value, the stronger the correlation. The sliding time window is a fixed time period, such as a 10-minute data analysis interval, which slides and updates over time. The feedforward compensation amount is a correction value based on prediction or historical data, which is used to adjust the replenishment amount in advance and reduce the error caused by the system response delay.

[0030] Exemplarily, the dynamic correlation weight matrix is an n-order square matrix, where the element w ij =α*(C ij / C max )+β*(T ij / T total ), α and β are the coefficients of the corresponding terms, α + β = 1, C ij represents the number of times food items i and j are jointly selected within the sliding time window, C max is the maximum number of joint selections, T ij is the selection time interval between the two food items, T total is the total window duration. The update period of the dynamic weight matrix is strictly synchronized with the sampling period of the proportional-integral control algorithm to ensure that the latest weight matrix is loaded at the beginning of each control cycle.

[0031] The actuator 12 is used to drive the replenishment robotic arm in the actuator to perform the replenishment operation according to the replenishment amount, the feedforward compensation amount and the pulse width modulation signal through the industrial bus protocol, and feedback and update the corresponding inventory data through the multi-modal sensing unit to form a closed-loop control loop.

[0032] The pulse width modulation signal is an electrical signal that controls the motor speed or actuator force by adjusting the duty cycle. The following is a specific example: During the lunch peak period in the cafeteria, the load cell monitors that the inventory weight of braised pork decreases by 150 g per minute. The RFID reader identifies that 80% of the plates pick up both braised pork and potatoes at the same time, and the visual recognition captures that the two are often picked up continuously. The coupling control unit generates an association weight of 0.7 between braised pork and potatoes, and predicts a real-time consumption rate of 180 g / min. The compensator shortens the integral time constant by 20% and calculates the replenishment amount as 200 g; the suppressor predicts the demand for the next 5 minutes and adds a feedforward compensation amount of 50 g; the trigger generates a pulse width modulation signal with a duty cycle of 60% because the rate exceeds the threshold. The actuator replenishes 200 g first, and then adds 50 g to complete the replenishment. After the inventory data is updated, it enters the next control cycle.

[0033] In the embodiment of the present application, the multi-modal sensing unit is used to collect the meal consumption data in real time. Combining the dynamic weight matrix and the closed-loop control algorithm, it realizes accurate prediction of the consumption rate and dynamic adjustment of the replenishment strategy. The compensator optimizes the integral time constant to cope with the collaborative consumption fluctuations. The suppressor responds to future demands in advance through feedforward compensation. The trigger quickly replenishes meals during sudden consumption. The actuator ensures the replenishment efficiency through priority scheduling. Finally, it reduces food waste, avoids insufficient supply, and improves the operation efficiency of the cafeteria.

[0034] In a possible embodiment, the coupling control unit includes a weight matrix generation module and a real-time consumption rate prediction module connected in sequence; the coupling control unit is used to establish a dynamic association weight matrix through the real-time data stream of the multi-modal sensing unit, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the replenishment amount deviation of the actuator. The dynamic association weight matrix is used to characterize the collaborative consumption characteristics between meals within the closed-loop control cycle, including: The weight matrix generation module is used to generate a dynamic weight matrix through the analysis of the dynamic coupling relationship within a sliding time window based on the real-time collected meal identification data and dynamic consumption association data, where the update period of the dynamic weight matrix is consistent with the sampling period of the proportional-integral control algorithm.

[0035] Among them, the sliding time window is a data analysis interval with a fixed duration (such as 10 minutes), which slides and updates as the system runs. The dynamic coupling relationship analyzes and quantifies the statistical correlation of different meals being picked up collaboratively within the same time window. The dynamic weight matrix is a normalized and encapsulated weight matrix, which can be adjusted in real time to adapt to changes in the device operation state and is used to optimize the control strategy. The real-time consumption rate prediction module is used to calculate the instantaneous consumption rate within the current time window according to the gradient change of the inventory weight change data, and perform a convolution operation on the instantaneous consumption rate and the dynamic weight matrix based on the error transfer function of the proportional-integral controller to obtain the real-time consumption rate.

[0036] Among them, the error transfer function is the mathematical model of the proportional-integral controller, which takes the input error signal and outputs the control quantity. The convolution operation is to perform weighted summation on the matrix and scalar according to the kernel function, and is used for data fusion and correction.

[0037] The following is a specific example: During the lunch peak period, the load cell monitors that the inventory of braised pork decreases by 200g per minute. The RFID reader records that 80% of the dinner plates take braised pork and rice at the same time, and the visual recognition device captures that the two are often taken continuously. The weight matrix generation module generates a dynamic weight matrix through sliding window analysis. The real-time consumption rate prediction module convolves the instantaneous rate of 200g / min with the weight matrix and outputs a corrected rate of 220g / min. The proportional-integral controller adjusts the replenishment quantity according to this rate, triggers the actuator to supplement 250g of braised pork, including a basic quantity of 200g and an associated compensation of 50g. At the same time, the replenishment quantity of rice increases by 40% synchronously. The system continuously optimizes the weight and rate prediction through closed-loop updates.

[0038] In the embodiment of the present application, through the dynamic weight matrix generation module, the system accurately quantifies the collaborative consumption relationship between food items. The real-time consumption rate prediction module combines the error transfer function of the PI controller to correct the instantaneous consumption rate into a real-time rate reflecting the influence of associated dishes.

[0039] In a possible embodiment, the weight matrix generation module includes a data processing sub-module, a timestamp recording sub-module, a hardware accelerator, and a matrix operation sub-module; the weight matrix generation module is used to generate a dynamic weight matrix based on the food item identification data and dynamic consumption association data collected in real time through the analysis of the dynamic coupling relationship within a sliding time window, including: The data processing sub-module is used to group and cache the food item identification data within the same meal selection combination based on the length of the time window, and generate a dynamic consumption association event data packet.

[0040] Among them, the dynamic consumption association event data packet is a data structure for recording the associated dish combinations within the same time window, and includes fields such as dish name pairs and combination frequencies.

[0041] The timestamp recording sub-module is used to add a corresponding timestamp to each dynamic consumption association event data packet, and trigger the incremental update of the coupling coefficient when the timestamp interval is less than the preset control period. Among them, the incremental update of the coupling coefficient is a coefficient that dynamically adjusts the association strength between dishes according to the time density.

[0042] A hardware accelerator is used to perform real-time encoding on meal identification data through a parallel computing architecture, generate meal feature vectors, and dynamically update the combined feature vectors based on the sampling period of the multimodal sensing unit. Among them, the parallel computing architecture utilizes the parallel computing units of field-programmable gate arrays or graphics processors to achieve high-speed feature encoding. The combined feature vector is a multi-dimensional vector formed by fusing the operating features of multiple devices, such as temperature, pressure, rotational speed, etc., and is used to characterize the comprehensive state of the device.

[0043] Exemplarily, the hardware accelerator is implemented using an existing chip, can include 64 parallel processing units, each processing unit is configured with 128KB of dedicated cache, interacts with the main processor through an existing bus, and the feature vector encoding delay can be less than 2ms.

[0044] A matrix operation sub-module is used to perform a tensor product operation on the combined feature vector and historical consumption rate data to generate a dynamic weight matrix that conforms to the industrial fieldbus data frame structure.

[0045] Among them, the tensor product operation is a vector outer product operation, which is used to map features and historical data into a matrix relationship. The industrial fieldbus data frame conforms to the standardized data encapsulation format of the industrial communication protocol. The historical consumption rate data records the resources of the device over a certain period of time in the past, such as the time series data of the power consumption rate and raw material consumption rate. The following is a specific example: During the lunch peak period, the plate radio frequency reader detects that 3 plates are taken with braised pork and rice simultaneously. The data processing sub-module generates an event packet. The timestamp recording sub-module finds that the interval between two events is only 0.8 seconds, less than the 1-second threshold, and increases the coupling coefficient of braised pork and rice from 0.7 to 0.75. The hardware accelerator encodes braised pork [1,0] and rice [0,1] into a combined vector [1,0,0,1]. The matrix operation sub-module performs a tensor product on it with the historical rate [200,150] to generate an initial matrix [[200,0,0,200],[150,0,0,150]]. After normalization, it is encapsulated into a data frame and input to the actuator to control the supplementary meal quantity to increase by 20% synchronously.

[0046] In the embodiment of this application, through the hardware accelerator and parallel computing, the real-time processing requirements of high-frequency data (once per second) are met. The timestamp triggers incremental updates to ensure that the weight matrix is quickly adjusted according to the actual consumption behavior. The data frame encapsulation enables the weight matrix to directly drive industrial actuators.

[0047] In a possible embodiment, a real-time consumption rate prediction module is configured to calculate an instantaneous consumption rate within a current time window based on the gradient change of inventory weight change data, and perform a convolution operation on the instantaneous consumption rate and a dynamic weight matrix using an error transfer function based on a proportional-integral controller to obtain a real-time consumption rate, including: Design a convolution kernel function based on the error transfer function of the proportional-integral controller. The convolution kernel function is a weight vector designed based on the PI controller parameters and is used for data fusion and correction.

[0048] Use the convolution kernel function to perform a discrete convolution operation on the instantaneous consumption rate and the dynamic weight matrix to obtain a real-time consumption rate. The real-time consumption rate and the deviation of the replenishment amount jointly adjust the proportional gain and integral gain of the error transfer function. The discrete convolution operation is to sum the matrix elements weighted by the kernel function to achieve data fusion. The deviation of the replenishment amount is the difference between the target replenishment amount and the real-time consumption rate, which drives the adjustment of the control parameters.

[0049] The following is a specific example: During the dinner peak period, the weighing sensor detects that the instantaneous consumption rate of braised pork is 200 g / min, and the dynamic weight matrix shows that the correlation weight between braised pork and rice is 0.8. Based on the current PI parameters, the proportional gain is 1.2 and the integral gain is 0.5, the system generates a convolution kernel [1.2, 0.05], performs a convolution operation on the instantaneous rate and the weight matrix, and outputs a real-time rate of 220 g / min. The deviation of 30 g from the target replenishment amount of 250 g triggers an increase in the proportional gain to 1.5 and an increase in the integral gain to 0.6. The updated convolution kernel [1.5, 0.06] is used for the next cycle calculation to make the replenishment amount more accurately match the actual consumption demand.

[0050] The embodiment of the present application integrates the dish correlation into the consumption rate prediction through convolution operation, improving the accuracy. The real-time rate and the deviation dynamically adjust the PI parameters, optimizing the replenishment response speed and stability. The adjusted PI parameters are fed back to the convolution kernel design to form a continuously improving control loop.

[0051] In a possible embodiment, a matrix operation sub-module is configured to perform a tensor product operation on a combined feature vector and historical consumption rate data to generate a dynamic weight matrix conforming to the industrial fieldbus data frame structure, including: The matrix operation sub-module is configured to perform a tensor product operation on the combined feature vector and historical consumption rate data to generate an initial weight matrix; the initial weight matrix is an unnormalized weight matrix generated through the tensor product operation, reflecting the preliminary weight distribution between the combined feature vector and the historical consumption rate data. The matrix operation sub-module is also used to normalize the initial weight matrix, encapsulate the normalized weight matrix in the data frame format of the industrial fieldbus protocol, and obtain the dynamic weight matrix. Among them, the industrial fieldbus protocol is a standardized protocol for communication between industrial devices, which defines data frame structures, transmission rules, etc. The data frame format is a data structure organized according to a specific protocol, usually including fields such as a frame header, data payload, and check bits to ensure the integrity and correctness of data transmission.

[0052] The following is a specific example: A certain enterprise cafeteria successfully optimized the supply of food during the lunch peak period by deploying a dynamic weight matrix system. The system first collected the current environmental characteristics: the coefficient of weekday lunchtime is 0.8, the coefficient of heavy rain weather is 0.4, and the coefficient of a sharp increase in the demand for vegetarian dishes is 0.7, and generated a combined feature vector F = [0.8, 0.4, 0.7]; at the same time, it retrieved historical data: the average consumption of braised pork ribs is 85 servings, broccoli with garlic is 120 servings, and scrambled eggs with tomatoes is 95 servings, and formed a historical consumption vector H = [85, 120, 95].

[0053] Through the tensor product operation, a 3×3 initial weight matrix is generated. After row normalization, [[0.52, 0.28, 0.42], [0.41, 0.34, 0.38], [0.39, 0.31, 0.36]] is obtained and encapsulated into an industrial data frame that conforms to the protocol. When the vision system detects that the heavy rain has caused a 40% sudden drop in the number of dine-in customers, the central controller immediately fuses and calculates the dynamic weight matrix with the real-time dining queue data, and outputs precise instructions to reduce the supply of braised pork ribs by 30%, maintain the benchmark for broccoli with garlic, and increase the supply of scrambled eggs with tomatoes by 20%. Finally, the food waste rate on that day is reduced from 19% to 7.2%, while ensuring that the last employee arriving can still get a complete dish supply.

[0054] In the embodiment of this application, the matrix operation sub-module performs a tensor product operation on the combined feature vector and the historical consumption rate data to generate an initial weight matrix, and then through normalization processing and industrial fieldbus protocol encapsulation, a dynamic weight matrix is obtained. This solution can dynamically adjust device control parameters, improve resource utilization, and at the same time ensure reliable transmission of data in the industrial network, and is applicable to scenarios such as intelligent manufacturing and energy management. In a possible embodiment, the actuator further includes a task scheduling module; the actuator is used to perform a replenishment operation according to the replenishment amount, feedforward compensation amount, and pulse width modulation signal, including: The task scheduling module is used to perform multi-task scheduling according to the replenishment amount, feedforward compensation amount, and pulse width modulation signal to obtain a task scheduling result.

[0055] Among them, multi-task scheduling is for the replenishment robotic arm to allocate the execution order and resources among multiple replenishment tasks to improve efficiency and response speed. The replenishment robotic arm is used to perform corresponding replenishment operations according to the task scheduling result.

[0056] Among them, the replenishment robotic arm is a mechanical device that automatically executes replenishment operations, usually consisting of a servo motor, an end effector, and a control system. The servo motor is a motor that can precisely control position and speed, and is used to drive the movement of the robotic arm. The following is a specific example: During the lunch peak period in a smart cafeteria, precise meal replenishment is achieved through the task scheduling module. When the central prediction system detects that the real-time consumption rate of braised pork exceeds the expected value by 15%, it triggers the dynamic scheduling mechanism. The task scheduling module synchronously receives three groups of parameters: 1. The replenishment quantity matrix [braised pork +20 servings / 5 min, stir-fried seasonal vegetables ±0, mapo tofu -5 servings / 5 min], 2. The feedforward compensation quantity vector [+8, 0, -3], 3. The pulse width modulation signal PWM = 15 kHz. After calculation by the multi-objective optimization algorithm, a scheduling instruction is generated: prioritize the execution of the braised pork replenishment task, synchronously adjust the pulse interval of the conveyor belt stepping motor to 50 ms, and drive the six-axis replenishment robotic arm to complete 8 rounds of precise feeding at a constant speed of 0.18 m / s. During this period, the torque feedback is used to dynamically compensate for the offset of the end effector caused by the stacked plates. Finally, the emergency replenishment is completed within 7 minutes and 32 seconds, ensuring that there is no shortage of dishes during the window period, and at the same time avoiding 3.2 kg of food waste caused by over-replenishment, improving the meal supply continuity by 38% compared with the traditional fixed-time replenishment mode. In the embodiment of the present application, the task scheduling module dynamically optimizes the replenishment strategy, combines feedforward compensation to reduce the lag error, and uses the PWM signal to precisely control the movement of the robotic arm to achieve efficient and precise automatic replenishment. The replenishment robotic arm executes operations according to the scheduling result, and uses the PID algorithm to ensure stability, which is applicable to the replenishment scenario during the lunch peak period in a smart cafeteria. In a possible embodiment, the task scheduling module is used to perform multi-task scheduling based on the replenishment quantity, feedforward compensation quantity, and pulse width modulation signal to obtain the task scheduling result, including: The task scheduling module is used to respectively assign corresponding priority coefficients to the replenishment quantity, feedforward compensation quantity, and pulse width modulation signal. Among them, the priority coefficient is a numerical value that quantifies the urgency of the task, and the higher the value, the higher the execution priority. The urgency of the task includes very urgent, relatively urgent, generally urgent, and not urgent. The calculation of the priority coefficient is a conventional technology in the art, and the present application will not elaborate here.

[0057] The task scheduling module is also used to sort in descending order according to the priority coefficients to generate the task scheduling result. Among them, generating the scheduling result is to output an ordered task queue for the actuator to execute in sequence. The descending order sorting arranges the tasks from largest to smallest according to the priority coefficients. For example, the PWM signal coefficient 4 > the replenishment quantity coefficient 3 > the feedforward compensation quantity coefficient 2.4.

[0058] The following is a specific example: At the end of the lunch peak, the system detects a 30% shortage in the inventory of braised pork in brown sauce, predicts an increase in demand in the next 5 minutes, and triggers a PWM signal due to a sudden queue. The task scheduling module queues the PWM signal at the top of the queue, followed by the meal replenishment quantity, and finally the feedforward compensation quantity. The meal replenishment robotic arm first replenishes 200g of braised pork at a high speed with an 80% duty cycle, then performs meal replenishment at the standard speed of 50g, and finally completes the feedforward compensation. If a new PWM signal arrives during this period, the current task response is immediately interrupted.

[0059] In the embodiment of this application, by calculating the coefficient in real time, it is ensured that high-frequency tasks occur suddenly are executed first. The orderly scheduling avoids the action conflict of the robotic arm and improves the meal replenishment efficiency. High-priority tasks can interrupt low-priority tasks to meet emergency needs.

[0060] Figure 2 It is a flowchart of an intelligent regulation method for the supply quantity of food in a smart cafeteria provided by an embodiment of this application, which is applied to a computing device in an intelligent regulation system for the supply quantity of food in a smart cafeteria, such as Figure 2 shown, the method includes: S21. Obtain the real-time data stream of the multi-modal sensing unit through the coupling control unit, establish a dynamic correlation weight matrix, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the deviation of the meal replenishment quantity of the actuator. The dynamic correlation weight matrix is used to characterize the collaborative consumption characteristics of food items within the closed-loop control cycle.

[0061] S22. The compensator performs real-time compensation on the integral time constant of the proportional-integral control algorithm according to the dynamic correlation weight matrix, and calculates the required meal replenishment quantity.

[0062] S23. The suppressor calculates the feedforward compensation quantity within the future time window based on the inventory weight change data and the dynamic consumption rate fluctuation data within the sliding time window.

[0063] S24. The trigger generates a pulse width modulation signal whose duty cycle is positively correlated with the consumption rate gradient when the real-time consumption rate exceeds the preset consumption rate threshold.

[0064] S25. Send the meal replenishment quantity, the feedforward compensation quantity, and the pulse width modulation signal to the actuator, so that the actuator drives the meal replenishment robotic arm in the actuator to perform meal replenishment operations according to the meal replenishment quantity, the feedforward compensation quantity, and the pulse width modulation signal, and updates the corresponding inventory data through the multi-modal sensing unit feedback to form a closed-loop control loop.

[0065] Figure 2A method for intelligent regulation of the supply quantity of food in a smart cafeteria described in the illustrated embodiment has a similar implementation principle and technical effect to the intelligent regulation system for the supply quantity of food in a smart cafeteria, which will not be elaborated here.

[0066] In a possible design, Figure 1 A computing device in an intelligent regulation system for the supply quantity of food in a smart cafeteria according to the illustrated embodiment, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32. The processing component 32 includes a coupled control unit and a control module group.

[0067] The storage component 31 stores one or more computer instructions, where the one or more computer instructions are called and executed by the processing component 32.

[0068] The processing component 32 is configured to: obtain the real-time data stream of the multimodal sensing unit through the coupled control unit, establish a dynamic association weight matrix, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the replenishment quantity deviation of the actuator. The dynamic association weight matrix is used to characterize the collaborative consumption characteristics of food items within the closed-loop control cycle; compensate the integral time constant of the proportional-integral control algorithm in real time according to the dynamic association weight matrix through a compensator, and calculate the required replenishment quantity; calculate the feedforward compensation quantity within the future time window based on the inventory weight change data and the dynamic consumption rate fluctuation data within the sliding time window through a suppressor; generate a pulse width modulation signal with a duty cycle positively correlated with the consumption rate gradient when the real-time consumption rate exceeds a preset consumption rate threshold through a trigger; send the replenishment quantity, the feedforward compensation quantity, and the pulse width modulation signal to the actuator, so that the actuator drives the replenishment robotic arm in the actuator to perform a replenishment operation according to the replenishment quantity, the feedforward compensation quantity, and the pulse width modulation signal, and updates the corresponding inventory data through the multimodal sensing unit to form a closed-loop control loop.

[0069] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above methods.

[0070] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disks or optical discs.

[0071] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0072] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0073] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0074] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component, storage component, etc. may be basic server resources leased or purchased from a cloud computing platform.

[0075] The embodiments of the present application also provide a computer storage medium storing a computer program, and when the computer program is executed by a computer, the following Figure 1 intelligent control method for the supply quantity of meals in a smart cafeteria shown in the embodiments can be implemented.

[0076] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0077] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

[0078] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A smart canteen food supply intelligent control system, characterized by: include: A multimodal sensing unit, comprising a weighing sensor located at the bottom of the food container, a radio frequency reader / writer for the food plate, and a visual recognition device, is used to collect real-time data on food inventory weight changes, food identification data, and dynamic consumption association data, which characterizes the temporal relationship of coordinated consumption between food items. The computing device includes a coupling control unit and a control module group. The coupling control unit is used to establish a dynamic correlation weight matrix based on the real-time data stream of the multimodal sensing unit, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the deviation of the supplementary meal amount of the actuator. The dynamic correlation weight matrix is used to characterize the collaborative consumption characteristics of the food items within the closed-loop control cycle; The control module group includes a compensator, a suppressor and a trigger; The compensator is used to compensate the integral time constant of the proportional-integral control algorithm in real time according to the dynamic correlation weight matrix and calculate the required meal supplement amount; The suppressor is configured to calculate a feedforward compensation amount within a future time window based on the inventory weight change data and the dynamic consumption rate fluctuation data within the sliding time window; The trigger is configured to generate a pulse width modulation signal having a duty cycle positively correlated with a consumption rate gradient when the real-time consumption rate exceeds a preset consumption rate threshold; The actuator is used to drive the meal replenishment robot arm in the actuator to perform the meal replenishment operation based on the meal replenishment amount, the feedforward compensation amount and the pulse width modulation signal through the industrial bus protocol, and update the corresponding inventory data through the multimodal sensing unit feedback to form a closed-loop control loop.

2. The system according to claim 1, wherein: The coupling control unit includes a weight matrix generation module and a real-time consumption rate prediction module connected in sequence; The coupling control unit is used to establish a dynamic correlation weight matrix using the real-time data stream of the multimodal sensing unit, and calculate the real-time consumption rate based on the gradient feedback signal of the inventory weight change and the deviation of the actuator's meal replenishment amount. The dynamic correlation weight matrix is used to characterize the collaborative consumption characteristics of different meal items within the closed-loop control cycle, including: The weight matrix generation module is configured to generate a dynamic weight matrix based on the real-time collected meal identification data and dynamic consumption association data by analyzing the dynamic coupling relationship within a sliding time window, wherein the update period of the dynamic weight matrix is consistent with the sampling period of the proportional-integral control algorithm; The real-time consumption rate prediction module is used to calculate the instantaneous consumption rate in the current time window based on the gradient change of the inventory weight change data, and perform a convolution operation on the instantaneous consumption rate and the dynamic weight matrix based on the error transfer function of the proportional-integral controller to obtain the real-time consumption rate.

3. The system according to claim 2, characterized in that The weight matrix generation module includes a data processing submodule, a timestamp recording submodule, a hardware accelerator and a matrix operation submodule; The weight matrix generation module is used to generate a dynamic weight matrix based on the real-time collected meal identification data and dynamic consumption association data by analyzing the dynamic coupling relationship within a sliding time window, including: The data processing submodule is used to group and cache the meal identification data in the same meal selection combination based on the length of the time window, and generate a dynamic consumption association event data packet; The timestamp recording submodule is used to add a corresponding timestamp to each of the dynamic consumption associated event data packets, and trigger an incremental update of the coupling coefficient when the timestamp interval is less than a preset control period; The hardware accelerator is configured to encode the food identification data in real time through a parallel computing architecture to generate a food feature vector, and dynamically update the combined feature vector based on a sampling period of the multimodal sensing unit; The matrix operation submodule is used to perform a tensor product operation on the combined feature vector and the historical consumption rate data to generate a dynamic weight matrix that conforms to the industrial field bus data frame structure.

4. The system according to claim 2, wherein: The real-time consumption rate prediction module is configured to calculate the instantaneous consumption rate within the current time window based on the gradient change of the inventory weight change data, and perform a convolution operation based on the error transfer function of the proportional-integral controller on the instantaneous consumption rate and the dynamic weight matrix to obtain the real-time consumption rate, including: Design the convolution kernel function based on the error transfer function of the proportional-integral controller; The instantaneous consumption rate is discretely convolved with the dynamic weight matrix using a convolution kernel function to obtain a real-time consumption rate. The real-time consumption rate and the meal supplement deviation are used together to adjust the proportional gain and integral gain of the error transfer function.

5. The system according to claim 3, wherein: The matrix operation submodule is used to perform a tensor product operation on the combined feature vector and the historical consumption rate data to generate a dynamic weight matrix that conforms to the industrial field bus data frame structure, including: The matrix operation submodule is used to perform a tensor product operation on the combined feature vector and the historical consumption rate data to generate an initial weight matrix; The matrix operation submodule is further used to normalize the initial weight matrix, and encapsulate the normalized weight matrix according to the data frame format of the industrial field bus protocol to obtain a dynamic weight matrix.

6. The system according to claim 1, wherein: The executor also includes a task scheduling module; The actuator is used to perform a meal supplement operation according to the meal supplement amount, the feedforward compensation amount and the pulse width modulation signal, including: The task scheduling module is used to perform multi-task scheduling according to the meal supplement amount, the feedforward compensation amount and the pulse width modulation signal to obtain a task scheduling result; The meal replenishment robot arm is used to perform corresponding meal replenishment operations according to the task scheduling results.

7. The system according to claim 6, characterized in that The task scheduling module is used to perform multi-task scheduling according to the meal supplement amount, the feedforward compensation amount and the pulse width modulation signal to obtain a task scheduling result, including: The task scheduling module is used to assign corresponding priority coefficients to the meal supplement amount, feedforward compensation amount and pulse width modulation signal respectively; The task scheduling module is further used to sort the tasks in descending order according to the priority coefficients to generate a task scheduling result.

8. A method for intelligently controlling the supply of food in a smart cafeteria, characterized in that: The computing devices used in the smart canteen food supply intelligent control system include: The real-time data stream from the multimodal sensing unit is acquired through a coupled control unit, and a dynamic correlation weight matrix is established. The real-time consumption rate is calculated based on the gradient feedback signal of the inventory weight change and the deviation of the actuator's meal replenishment quantity. The dynamic correlation weight matrix is used to characterize the collaborative consumption characteristics of different meal items within the closed-loop control cycle. The compensator compensates the integral time constant of the proportional-integral control algorithm in real time according to the dynamic correlation weight matrix and calculates the required meal supplement amount; The suppressor calculates the feedforward compensation amount in the future time window based on the inventory weight change data and dynamic consumption rate fluctuation data in the sliding time window; When the real-time consumption rate exceeds a preset consumption rate threshold, a trigger is used to generate a pulse width modulation signal having a duty cycle positively correlated with a consumption rate gradient; The meal replenishment amount, the feedforward compensation amount and the pulse width modulation signal are sent to the actuator, so that the actuator drives the meal replenishment robot arm in the actuator to perform the meal replenishment operation through the industrial bus protocol according to the meal replenishment amount, the feedforward compensation amount and the pulse width modulation signal, and updates the corresponding inventory data through the multimodal sensing unit feedback to form a closed-loop control loop.

9. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a smart canteen food supply intelligent control method as described in claim 8.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for intelligently controlling the supply quantity of food in a smart canteen as claimed in claim 8 is implemented.

Images