On-off control method and system for household stirrer

Abstract

The invention relates to a switch control method and system for a household stirrer, and relates to the technical field of digital control, and the method comprises the steps: obtaining the food weight and food type of to-be-stirred target food through a weight sensor in the stirrer, carrying out the stirring time prediction, and obtaining the predicted stirring time; in the stirring starting process, monitoring and recording motor power to obtain a motor power sequence; processing and acquiring a time coefficient of the accumulated stirring time and the predicted stirring time, analyzing and acquiring a power change coefficient of a motor power sequence, and configuring a stirring state recognition coefficient; according to the stirring state recognition coefficient, stirring state recognition is conducted based on the motor power sequence and the accumulated stirring time, stirring state information is obtained, and when the preset stirring state is met, the stirrer is closed. The technical problem that in the prior art, a household stirrer is not intelligently controlled, and the use effect of the household stirrer is affected is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital control, and particularly to a switch control method and system for a household blender. Background Art

[0002] Household blenders are commonly used for stirring foods such as cream and eggs. Generally, they perform stirring through constant speed control, and the user judges whether the stirring is completed based on experience and then conducts switch control.

[0003] Therefore, the switch control of the household blender in the prior art is not intelligent enough. It is possible that the stirring state of the food does not meet the required standards when the user turns it off, and it is necessary to stir again. There are technical problems such as non-intelligent switch control and affecting the use effect of the household blender. Summary of the Invention

[0004] In view of the technical problems of non-intelligent switch control of the household blender and affecting the use effect of the household blender in the prior art, the present invention provides a switch control method and system for a household blender.

[0005] The technical solutions for the present invention to solve the above technical problems are as follows:

[0006] In a first aspect, the present invention provides a switch control method for a household blender, including: obtaining the food weight and food type of the target food to be stirred through a weight sensor in the blender, predicting the stirring time, and obtaining the predicted stirring time;

[0007] During the start of stirring, monitoring and recording the motor power to obtain a motor power sequence;

[0008] Processing to obtain the time coefficient of the cumulative stirring time and the predicted stirring time, and analyzing to obtain the power change coefficient of the motor power sequence, and configuring a stirring state recognition coefficient;

[0009] Based on the stirring state recognition coefficient, performing stirring state recognition based on the motor power sequence and the cumulative stirring time to obtain stirring state information, and turning off the blender when the preset stirring state is met.

[0010] In a second aspect, the present invention provides a switch control system for a household blender, including:

[0011] A stirring time prediction module for obtaining the food weight and food type of the target food to be stirred through a weight sensor in the blender, predicting the stirring time, and obtaining the predicted stirring time;

[0012] A motor power monitoring module for monitoring and recording the motor power during the start of stirring to obtain a motor power sequence;

[0013] A stirring recognition configuration module, which is used to process and obtain the time coefficient of the cumulative stirring time and the predicted stirring time, analyze and obtain the power change coefficient of the motor power sequence, and configure the stirring state recognition coefficient.

[0014] An identification switch control module, which is used to identify the stirring state based on the motor power sequence and the cumulative stirring time according to the stirring state recognition coefficient, obtain the stirring state information, and turn off the blender when the preset stirring state is satisfied.

[0015] The beneficial effects of the present invention are as follows: The present invention provides a switch control method and system for a household blender. Compared with the traditional manual control of the switch relying on user experience, it can realize intelligent stirring switch control based on the food weight, food type, and the change characteristics of the motor power, improve the automation degree of stirring and the food stirring effect, and further improve the user experience. It can accurately judge the self-start of the blender by obtaining the food weight through a weight sensor and combining the food type. The stirring time prediction method based on historical data makes the stirring time more accurate and avoids insufficient or over-stirring caused by human judgment errors. During the stirring process, by monitoring the change of the motor power to form a motor power sequence and combining the cumulative stirring time, the stirring state can be evaluated in real time. Compared with the traditional method that only relies on fixed-time control, it can better adapt to the stirring requirements of different foods and improve the stirring quality. In addition, by calculating the time coefficient and the power change coefficient to construct the stirring state recognition coefficient, it can more accurately identify the actual stirring state of the food and judge when to stop stirring, avoid repeatedly turning on the blender by humans to adjust the stirring degree, improve the use convenience, and optimize the user experience. Through multi-dimensional data analysis and automatic control, the present invention makes the switch control of the household blender more intelligent, improves the stirring efficiency and quality, and solves the technical problems of inaccurate judgment of the stirring state and non-intelligent switch control of the blender in the prior art. Description of the Drawings

[0016] Figure 1 It is a schematic flow chart of a switch control method for a household blender provided by the present invention.

[0017] Figure 2 It is a schematic structural diagram of a switch control system for a household blender provided by the present invention.

[0018] Reference numerals: stirring time prediction module 11, motor power monitoring module 12, stirring recognition configuration module 13, identification switch control module 14. Detailed Embodiments

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0020] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, the meaning of "a plurality" is two or more, unless otherwise specifically defined.

[0021] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for the purpose of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present invention.

[0022] Embodiment 1, as Figure 1 shown, the embodiment of the present invention provides a switch control method for a household blender, and the method specifically includes the following steps:

[0023] S10: Through the weight sensor in the blender, obtain the food weight and food type of the target food to be stirred, predict the stirring time, and obtain the predicted stirring time.

[0024] In the embodiments of the present application, the present invention detects the food to be stirred through a weight sensor (a sensor element for detecting food weight, such as a strain gauge sensor or a piezoelectric sensor) pre-placed in the blender to obtain its food weight.

[0025] Exemplarily, the weight sensor can be pre-set inside the bottom of the blender, and weight sensing acquisition is performed in the standby state when the blender is not started to obtain the food weight.

[0026] Further, to ensure that the food stirring state meets the requirements, for example, the cream reaches the whipped state, according to the food weight and food type, predict the stirring time until completion, and obtain the predicted stirring time. Among them, the food type can be obtained by the user inputting on the menu selection interface of the blender.

[0027] Step S10 in the method provided by the embodiment of the present application includes:

[0028] Obtain the food weight of the target food to be stirred through the weight sensor in the blender;

[0029] Receive the food type input by the user.

[0030] In the embodiment of the present application, the weight sensor in the blender (a sensing device for measuring the food weight in the container, such as a strain gauge sensor, a piezoelectric sensor or a capacitive sensor, is configured at the bottom of the blender container to test the food weight) weighs the target food to be stirred to obtain the food weight.

[0031] The target food is the food raw material placed in the blender container and about to be stirred, such as egg white, cream, batter, etc. The food weight is the mass of the target food, in grams (g). For example, the measured cream weight is 300 g and the batter weight is 500 g. The weight sensor can monitor the food weight in real time and transmit the food quality data to the processing chip of the blender through data transmission for subsequent processing. The blender in the embodiment of the present application executes the method through the built-in chip and can be connected to the network through the Internet of Things through the network connection device (such as a Wi-Fi module, a Bluetooth module or an NB-IoT communication module), and executes the method through the network connection server.

[0032] While detecting the weight, receive the food type input by the user (that is, the food category information provided by the user, such as "egg white", "cream" or "batter"). The input method can be that the user manually selects the preset food category on the operation interface of the blender. For example, the user can select the food category of "egg white" on the touch screen of the blender, so as to obtain the food type.

[0033] Different food types and food weights require different stirring times. By combining the information of food weight and food type, an appropriate stirring strategy can be intelligently matched to achieve more precise control and improve the quality and uniformity of food stirring.

[0034] Step S10 in the method provided by the embodiment of the present application further includes: collecting a sample food weight set and a sample food type set according to the stirring historical data of the same type of blender;

[0035] Collect the time required to reach the required stirring state when stirring different sample food types and different sample food weights, and label it as the sample stirring time set;

[0036] Construct the mapping relationship of the sample food weight set, the sample food type set, and the sample stirring time set to obtain a stirring time prediction table;

[0037] Input the food weight and food type into the stirring time prediction table, and output the predicted stirring time.

[0038] In the embodiments of the present application, the mixer stirs at a constant speed, and a prediction table for predicting the stirring time is constructed based on the stirring historical data of the same type of mixer (i.e., the stirring data of the same model mixer during use and testing).

[0039] First, based on the stirring historical data of the same type of mixer, the sample food weights and sample food types collected when the same type of mixer stirs, obtain the sample food weight set and the sample food type set.

[0040] Furthermore, collect the time required to reach the required stirring state when stirring different sample food types and different sample food weights, and label this time data as the sample stirring time set (i.e., the time required for the food to be stirred to an ideal state, for example, it takes 3 minutes to whip 200g of egg white to a stiff peak, and 5 minutes to stir 500g of batter to be uniform and delicate). The required stirring state can be determined by testing the density of different types of food after stirring to reach the requirement. For example, air is incorporated into the cream during stirring, and the density will decrease. By testing the time when its density reaches the whipped requirement density, the sample stirring time can be obtained.

[0041] Furthermore, construct the mapping relationship of the sample food weight set, the sample food type set, and the sample stirring time set to form a stirring time prediction table. This stirring time prediction table can be used to predict the required stirring time based on the food weight and food type. Part of the content of this stirring time prediction table is shown in Table 1.

[0042] Sample food types Sample food weight Sample stirring time Egg white 200g 3 minutes Cream 300g 4.5 minutes Batter 500g 5 minutes

[0043] Table 1

[0044] Furthermore, input the current food weight and food type into the stirring time prediction table for mapping retrieval to obtain the predicted stirring time. Exemplarily, first map the data of the same type of food according to the food type, and then map the sample food weight corresponding to the same sample food weight or the closest (i.e., the smallest weight difference) sample food weight through the food weight as the predicted stirring time.

[0045] By using the stirring data within a historical time period to predict the stirring time for the current food weight and food type, it can provide a reference for controlling the stirring switch, ensure that the food stirring meets the requirements, and avoid the situation where stirring needs to be restarted due to insufficient stirring.

[0046] S20: During the process of starting stirring, monitor and record the motor power to obtain a motor power sequence.

[0047] In the embodiments of the present application, during the process of the user starting the household blender for stirring, the motor power is monitored and recorded in real time to form a motor power sequence, which is used to analyze the stirring state of the food and determine whether the stirring is completed.

[0048] Among them, during the food stirring process, due to differences between different batches of food, there may be errors in predicting the stirring time, resulting in the food being stirred completed before the predicted stirring time or the food still not being stirred completed after the predicted stirring time. Therefore, based on the motor power, it is analyzed whether the food stirring is completed.

[0049] Among them, the motor power of the blender refers to the electrical energy power consumed by the motor when stirring the food, which is usually determined by the current and voltage of the motor. During the process of the blender stirring the food at a constant speed, different food types and stirring stages will cause changes in the motor power. For example, when the batter starts to be stirred, it is relatively viscous. To ensure the rotation speed, the motor load is large and the power is high. As the batter gradually becomes uniform, the stirring resistance decreases, and at a constant rotation speed, the power will gradually tend to be stable. Therefore, the change in the motor power can reflect the stirring state of the food and is an important parameter for judging whether the stirring is completed.

[0050] The step S20 in the method provided by the embodiments of the present application includes:

[0051] During the stirring process, continuously monitor the motor power at the most recent multiple timestamps;

[0052] Arrange the motor power at the most recent multiple timestamps in chronological order to obtain a motor power sequence.

[0053] In the embodiments of the present application, during the process of the user starting the blender for stirring, continuously monitor the motor power at the most recent multiple timestamps and arrange these data in chronological order to form a motor power sequence, which is used to analyze the change in the food stirring state.

[0054] When the motor of the blender is working, its power consumption is affected by the food viscosity, density, and stirring uniformity. Therefore, real-time collection of the motor power data can reflect the state changes of the food at different stirring stages.

[0055] Specifically, through a current sensor (such as a Hall effect sensor), the working current and voltage of the mixer motor are collected during the mixing process, and the current motor power is calculated. This data collection process uses a fixed time interval, such as recording the power value every 100 ms or 500 ms, to ensure the continuity of the data. For example, during the cream mixing process, the power value may be recorded every 500 ms, thereby obtaining the motor power at multiple time stamps.

[0056] Further, to improve the processing efficiency, only the power data of the most recent multiple time stamps is retained, that is, a sliding window mechanism is adopted. This method can ensure that only the latest data is processed without storing the power data of the entire mixing process. For example, if the sliding window is set to the most recent 10 time points, then for a sampling interval of 500 ms, this window stores at most 10 motor power data within the past 5 seconds.

[0057] Further, all the monitored power data is arranged in chronological order to form a motor power sequence for subsequent analysis of the mixing state. For example, during the cream mixing process, due to the entry of air, the viscosity and density of the cream decrease, and the motor power sequence may show a downward trend.

[0058] In the embodiment of the present invention, by continuously monitoring the motor power of the most recent multiple time stamps and arranging them in chronological order to form a motor power sequence, accurate data support is provided for the analysis of the mixing state.

[0059] S30: Process and obtain the time coefficient of the cumulative mixing time and the predicted mixing time, analyze and obtain the power change coefficient of the motor power sequence, and configure the mixing state recognition coefficient.

[0060] In the embodiment of the present application, it is necessary to analyze whether the mixing state of the food meets the requirements based on the motor power sequence. Specifically, the mixer is connected to the cloud server through the Internet of Things, and the cloud server is used to identify the mixing state of the food. Among them, in order to improve the recognition response efficiency, reduce the computing power consumption and improve the recognition accuracy, the time coefficient is calculated and configured according to the cumulative mixing time during the mixing process, and the power change coefficient is configured according to the motor power sequence to configure the mixing state recognition coefficient for the use of the mixing state recognition computing power.

[0061] Step S30 in the method provided by the embodiment of the present application includes:

[0062] Monitor and obtain the cumulative mixing time after the mixer is started;

[0063] According to the cumulative mixing time and the predicted mixing time, process and obtain the time coefficient as follows:

[0064]

[0065] where KT is the time coefficient, T l is the cumulative stirring time, and T y is the predicted stirring time.

[0066] In the embodiment of the present application, after the mixer is started, based on the starting time node, the stirring working time of the motor is monitored in real time, that is, the cumulative stirring time. For example, when the mixer starts, the timer starts recording from 0 seconds. If the stirring lasts for 10 seconds, the cumulative time is 10 seconds.

[0067] According to the cumulative stirring time and the predicted stirring time, the time coefficient is obtained through processing, as shown in the following formula:

[0068]

[0069] where KT is the time coefficient, reflecting the proportion of computing power resources configured according to the cumulative stirring time, T l is the cumulative stirring time, and T y is the predicted stirring time.

[0070] The shorter the cumulative stirring time is, the greater the probability that the food stirring state has not reached the required stirring state, such as the density not meeting the standard. At this time, less computing power can be configured for the computing power to identify the food stirring state according to the motor power sequence, so as to save computing power consumption and improve data transmission and response efficiency. The greater the cumulative stirring time is, the greater the probability that the food stirring state meets the requirements, and more computing power resources are configured to improve the accuracy of identifying the food stirring state according to the motor power sequence. When the cumulative stirring time is greater than the predicted stirring time, the time coefficient is 1, and all computing power resources are configured to identify the food stirring state according to the motor power sequence, improving the recognition accuracy and the stirring quality.

[0071] Exemplarily, if the predicted stirring time is 180 seconds and the cumulative stirring time is 120 seconds, the time coefficient is 120 / 180 = 0.67.

[0072] Step S30 in the method provided by the embodiment of the present application further includes:

[0073] Randomly select two motor powers multiple times within the motor power sequence and calculate the power fluctuation amplitude;

[0074] Output the maximum power fluctuation amplitude as the power fluctuation coefficient;

[0075] Subtract the power fluctuation coefficient from 1 to obtain the power change coefficient;

[0076] Calculate the stirring state recognition coefficient according to the time coefficient and the power change coefficient.

[0077] In the embodiments of the present application, in addition to configuring the computing power resources for identifying the stirring state according to the cumulative stirring time based on the motor power sequence, the computing power resources for identifying the stirring state are also configured according to the change amplitude of the motor power within the motor power sequence. Among them, if the change amplitude of the motor power is large, it indicates that the food stirring state is not yet stable and is still in the process of change. Most likely, the stirring state does not meet the requirements. Then, less computing power resources are configured to identify the stirring state to save computing power resources. On the contrary, if the change amplitude of the motor power is small, it indicates that the food stirring state is stable. Most likely, the stirring state meets the requirements. Then, more computing power resources are configured to identify the stirring state to improve the recognition accuracy.

[0078] In the embodiments of the present application, two motor power values are randomly selected from the collected motor power sequence, and the power fluctuation amplitude between them is calculated. Exemplarily, two motor power values are randomly selected, the absolute value of the difference between the two is calculated, and the ratio of the absolute value to the smaller motor power value between the two is calculated as the power fluctuation amplitude.

[0079] Two motor powers are randomly selected multiple times, such as 3 times, to obtain 3 sets of motor powers. The power fluctuation amplitude is calculated, and the maximum power fluctuation amplitude is output as the power fluctuation coefficient.

[0080] Further, the greater the power fluctuation, the smaller the proportion of the computing power for identifying the stirring state configured based on the power fluctuation amplitude, that is, the smaller the power change coefficient. Therefore, 1 minus the power fluctuation coefficient is used to obtain the power change coefficient as the proportion of the computing power resources configured based on the power fluctuation.

[0081] Further, according to the two dimensions of the time coefficient and the power change coefficient, the computing power configuration ratio for finally identifying the stirring state, that is, the stirring state recognition coefficient, is calculated. Exemplarily, the mean value of the time coefficient and the power change coefficient is calculated as the final identification of the stirring state. Or the time coefficient and the power change coefficient are weighted and calculated with weights of 0.7 and 0.3 to obtain the final identification of the stirring state. There are errors in power monitoring, so the configured weight is small.

[0082] The embodiments of the present application configure the computing power resources for identifying the food stirring state based on the cumulative stirring time and the change of the motor power, which can save computing power consumption, improve the processing efficiency, and improve the recognition accuracy of the stirring state when the food is about to be stirred to completion, provide an accurate judgment of the working state for the blender, and intelligently control the end timing of the stirring process, significantly improving the stirring quality and user experience.

[0083] S40: Based on the motor power sequence and the cumulative stirring time, the stirring state is identified according to the stirring state recognition coefficient to obtain the stirring state information. When the preset stirring state is satisfied, the blender is turned off.

[0084] In the embodiments of the present application, computing resources for stirring state recognition are configured through a stirring state recognition coefficient. Specifically, the number of model branches for stirring state recognition is configured, and the stirring state is recognized based on the motor power sequence and the cumulative stirring time to obtain the actual stirring state information of the food, such as density. When the preset stirring state is met, the mixer is turned off, realizing the intelligent automatic shutdown of the mixer, improving the intelligence of the mixer switch control and the effect of food stirring, and enhancing the user experience.

[0085] Step S40 in the method provided by the embodiments of the present application includes:

[0086] In the cloud server, a stirring state recognizer including X stirring state recognition branches is trained, where X is a positive integer;

[0087] After the time coefficient is greater than or equal to the time coefficient threshold, Y is calculated and rounded by multiplying the stirring state recognition coefficient by X;

[0088] Randomly select Y stirring state recognition branches in the stirring state recognizer, transmit the motor power sequence to the cloud server and input it, and output to obtain the stirring state information of Y branches, and calculate to obtain the stirring state information;

[0089] Judge whether the stirring state information meets the preset stirring state. If so, turn off the mixer. If not, judge whether the cumulative saturated stirring time when the time coefficient is 1 is greater than the saturated stirring time threshold. If so, turn off the mixer. If not, continue stirring.

[0090] In the embodiments of the present application, in order to improve the accuracy and generalization ability of stirring state recognition, first, a stirring state recognizer including X stirring state recognition branches is trained on the cloud server. X is a positive integer, for example, 10. Among them, the computing resources of the mixer locally cannot run the stirring state recognizer, so it is configured to be performed on the cloud. The mixer itself transmits the motor power sequence to the cloud server through the Internet of Things for recognition, and then discriminant processing is performed, and the control signal is transmitted back to the mixer for control.

[0091] In the embodiments of the present application, in the cloud server, training a stirring state recognizer including X stirring state recognition branches includes:

[0092] According to the stirring data of the same food type in the historical time, collect the sample motor power sequence set, and collect the stirring state information of the food under different sample motor power sequences to obtain the sample stirring state information set;

[0093] Divide the sample motor power sequence set and the sample stirring state information set to obtain X stirring state recognition training data;

[0094] Use the X stirring state recognition training data respectively to train X stirring state recognition branches and obtain a stirring state recognizer.

[0095] In the embodiments of the present application, in order to achieve accurate recognition of the stirring state of food, a machine learning method based on historical stirring data is adopted to deeply correlate and model the motor power sequence and the food stirring state during the stirring process, and a stirring state recognizer is trained and obtained.

[0096] Among them, during the actual stirring process, the physical properties of different food types (such as cream, batter, egg white, etc.) are different, and their stirring states will affect the power consumption of the motor. In order to establish the mapping relationship between the stirring state and the power, different stirring state recognizers are trained based on different food types and configured to be used on the cloud server.

[0097] In the embodiments of the present application, according to the stirring experiment test data of the same type of mixer for the same food type within the historical time, the motor power sequence monitored and recorded during the stirring process is collected to obtain a sample motor power sequence set. Further, the stirring state information of the food under different sample motor power sequences, such as density, is collected to obtain a sample stirring state information set. For example, the stirring can be stopped under different sample motor power sequences, the height of the food in the mixer is measured, the volume is calculated according to the space inside the mixer, and then combined with the measured food weight, the food density is calculated as the sample stirring state information.

[0098] Exemplarily, the sample motor power sequence can be {250W, 260W, 255W, 245W, 240W, 238W, 236W}, and the sample stirring state information can be 0.90g / cm 3 , 0.75g / cm 3 etc.

[0099] Further, in order to ensure the accuracy of the stirring state recognizer, X stirring state recognition branches are trained by using ensemble learning. X is 10 for example. Through the training of multiple stirring state recognition branches, subsequent integrated stirring state recognition is carried out to ensure the recognition accuracy and generalization ability.

[0100] The sample motor power sequence set and the sample stirring state information set are divided, for example, into X equal parts respectively to form X groups of stirring state recognition training data. Each group of stirring state recognition training data includes multiple corresponding sample motor power sequences and sample stirring state information.

[0101] Further, use the X stirring state recognition training data respectively to train X stirring state recognition branches, and combine the X trained stirring state recognition branches to obtain a stirring state recognizer.

[0102] Exemplarily, X stirring state recognition branches in machine learning are trained using a feedforward neural network. Taking one of the stirring state recognition branches as an example, the network structure and training steps of the stirring state recognition branch are described. First, a feedforward neural network is used to construct the stirring state recognition branch, including an input layer, an output layer, and a hidden layer. Its input feature is the electric shock power sequence, and the input dimension of the input layer is 10, that is, the motor power within the last 10 timestamps. The output feature is the stirring state information (such as density), and the output dimension is 1. The output layer uses a linear activation function, and the hidden layer uses a ReLU activation function. The stirring state recognition training data is divided into a training set and a test set in a ratio of 8:2. The training set is used for training, and the test set is used for testing performance. During the training process, the sample motor power sequence is input. After being calculated by the hidden layer, the output layer generates the predicted stirring state information, such as the predicted density. Then, according to the true sample stirring state information, the mean square error loss function is used to calculate the error between the two as the loss. Then, the network parameters are adjusted by backpropagation. This step is repeated until the loss converges, for example, the error is less than 5%. Through iterative training and adjustment of multiple groups of training data, finally, testing is performed. If the test loss converges, the training is completed. In this way, the supervised training of X stirring state recognition branches can be completed.

[0103] After all X stirring state recognition branches are trained, a trained stirring state recognizer is obtained by combination. In the embodiment of the present application, by training the stirring state recognizer, with the motor power sequence as the input, the state information of the food during stirring, such as density, can be predicted, and it can automatically determine whether the stirring is completed, thereby optimizing the stirring time and effect and improving the automation degree of the blender. Through the machine learning method and combining the historical data in actual operation, not only can the accuracy of the stirring effect be improved, but also the human intervention can be reduced and the operation efficiency can be enhanced.

[0104] In the embodiment of the present application, the monitoring time coefficient is used to judge whether the stirring is about to be completed, and the stirring state recognition is performed when it is about to be completed to further save the use of computing power. Specifically, it is judged whether the time coefficient is greater than the time coefficient threshold. The time coefficient threshold is, for example, 0.5. After the time coefficient is greater than or equal to the time coefficient threshold, it means that the cumulative stirring time has reached half of the predicted stirring time, and it may have been stirred to completion, and the stirring state recognition is performed.

[0105] Specifically, the number Y of the selected stirring state recognition branches is obtained by multiplying the stirring state recognition coefficient by the total number X of the stirring state recognition branches and rounding up. In this way, the configuration of the computing power resources for the stirring state recognition is completed. Exemplarily, the stirring state recognition coefficient is 0.55, and Y is equal to 0.55*X and rounded up to be equal to 6. Y is a positive integer less than or equal to X.

[0106] The larger the stirring state recognition coefficient, the greater the probability of completing stirring, the more branches are called, and the higher the recognition accuracy. Conversely, the smaller the stirring state recognition coefficient, the smaller the probability of completing stirring, the fewer branches are called, and the computing power is saved.

[0107] Further, randomly select Y stirring state recognition branches in the stirring state recognizer, transmit the current motor power sequence to the cloud server, input the selected Y stirring state recognition branches, output and obtain the stirring state information of Y branches, calculate the mean value of the stirring state information of Y branches, and obtain the stirring state information, such as the predicted average density.

[0108] Judge whether the stirring state information meets the preset stirring state. For example, judge whether the density in the stirring state information is less than or equal to the preset density value when the food stirring is completed. If so, the stirring is completed and the mixer is turned off. Exemplarily, the preset density value for cream food is 0.4g / cm 3 , if the density in the recognized stirring state information is less than this preset density value, it means that the stirring is completed.

[0109] If not, in order to avoid continuous stirring caused by data acquisition and processing errors, further judge whether the cumulative saturated stirring time with a time coefficient of 1 is greater than the saturated stirring time threshold, that is, whether the time length of the cumulative stirring time exceeding the predicted stirring time is greater than the saturated stirring time threshold. If so, it means that the stirring time is sufficient and the stirring is stopped. If not, it means that the stirring is not completed yet and the stirring continues. The saturated stirring time threshold is, for example, 30S.

[0110] The embodiment of the present application optimizes the stirring process based on the time coefficient, power fluctuation characteristics, intelligent recognition and dynamic time judgment, ensures that the food reaches an ideal stirring state, performs automatic control of the mixer switch, improves the stirring efficiency at the same time, avoids over-stirring or under-stirring, and improves the automation and intelligence level.

[0111] Optionally, the mixer in the embodiment of the present application is also equipped with a manual switch control button, for example, a rotary switch button. When it is rotated to a predetermined position, the stirring is turned on. When it is not rotated to the predetermined position, it is turned off. By using the manual switch control button in combination with the method provided by the embodiment of the present application, diversified applications are realized and the user experience is improved.

[0112] A switch control method for a household mixer provided by an embodiment of the present invention has at least the following technical effects:

[0113] Compared with the traditional method of manually controlling the switch relying on user experience, the embodiment of the present invention can realize intelligent stirring switch control based on the food weight, food type, and motor power change characteristics, improve the automation degree of stirring and the food stirring effect, and further improve the user experience. It can accurately judge the self-start of the blender by obtaining the food weight through a weight sensor and combining the food type. The stirring time prediction method based on historical data makes the stirring time more accurate and avoids under-stirring or over-stirring caused by human judgment errors. During the stirring process, by monitoring the change of motor power to form a motor power sequence and combining the cumulative stirring time, the stirring state can be evaluated in real time. Compared with the traditional method that only relies on fixed-time control, it can better adapt to the stirring needs of different foods and improve the stirring quality. In addition, by calculating the time coefficient and power change coefficient, a stirring state recognition coefficient is constructed, which enables more accurate recognition of the actual stirring state of the food and judges when to stop stirring, avoiding repeated manual opening of the blender to adjust the stirring degree, improving the use convenience, and optimizing the user experience. Through multi-dimensional data analysis and automatic control, the present invention makes the switch control of the household blender more intelligent, improves the stirring efficiency and quality, and solves the technical problems of inaccurate judgment of the stirring state and non-intelligent switch control of the blender in the prior art.

[0114] Embodiment 2, as Figure 2 shown, with the same inventive concept as the switch control method for a household blender in Embodiment 1, the embodiment of the present invention further provides a switch control system for a household blender. The explanation of the switch control method for a household blender in Embodiment 1 is also applicable to a switch control system for a household blender. The system includes:

[0115] A stirring time prediction module 11, configured to obtain the food weight and food type of the target food to be stirred through a weight sensor in the blender, predict the stirring time, and obtain the predicted stirring time;

[0116] A motor power monitoring module 12, configured to monitor and record the motor power during the stirring start process to obtain a motor power sequence;

[0117] A stirring recognition configuration module 13, configured to process and obtain the time coefficient of the cumulative stirring time and the predicted stirring time, analyze and obtain the power change coefficient of the motor power sequence, and configure a stirring state recognition coefficient;

[0118] An identification switch control module 14, configured to perform stirring state identification based on the motor power sequence and the cumulative stirring time according to the stirring state recognition coefficient, obtain stirring state information, and turn off the blender when a preset stirring state is satisfied.

[0119] Further, the stirring time prediction module 11 is further configured to:

[0120] Obtain the food weight of the target food to be stirred through the weight sensor in the blender;

[0121] Receive the food type input by the user.

[0122] Furthermore, the stirring time prediction module 11 is also used for:

[0123] According to the stirring historical data of the same type of blender, collect the sample food weight set and the sample food type set;

[0124] Collect the time required to stir to the required stirring state under different sample food types and different sample food weights, and label it as the sample stirring time set;

[0125] Construct the mapping relationship among the sample food weight set, the sample food type set and the sample stirring time set to obtain the stirring time prediction table;

[0126] Input the food weight and food type into the stirring time prediction table, and output the predicted stirring time.

[0127] Furthermore, the motor power monitoring module 12 is also used for:

[0128] During the stirring process, continuously monitor the motor power at the most recent multiple timestamps;

[0129] Arrange the motor power at the most recent multiple timestamps in chronological order to obtain the motor power sequence.

[0130] Furthermore, the stirring recognition configuration module 13 is also used for:

[0131] Monitor and obtain the cumulative stirring time after the blender is started;

[0132] According to the cumulative stirring time and the predicted stirring time, process and obtain the time coefficient as follows:

[0133]

[0134] where, KT is the time coefficient, T l is the cumulative stirring time, T y is the predicted stirring time.

[0135] Furthermore, the stirring recognition configuration module 13 is also used for:

[0136] Randomly select two motor powers multiple times within the motor power sequence and calculate the power fluctuation amplitude;

[0137] Output the maximum power fluctuation amplitude as the power fluctuation coefficient;

[0138] Obtain the power change coefficient by subtracting the power fluctuation coefficient from 1;

[0139] Calculate and obtain the stirring state recognition coefficient according to the time coefficient and the power change coefficient.

[0140] Furthermore, the identification switch control module 14 is further configured to:

[0141] In the cloud server, train a stirring state recognizer including X stirring state recognition branches, where X is a positive integer;

[0142] After the time coefficient is greater than or equal to the time coefficient threshold, calculate and round down to obtain Y by multiplying the stirring state recognition coefficient by X;

[0143] Randomly select Y stirring state recognition branches in the stirring state recognizer, transmit the motor power sequence to the cloud server and input it, output to obtain the stirring state information of Y branches, and calculate to obtain the stirring state information;

[0144] Judge whether the stirring state information meets the preset stirring state. If so, turn off the mixer. If not, judge whether the cumulative saturated stirring time when the time coefficient is 1 is greater than the saturated stirring time threshold. If so, turn off the mixer. If not, continue stirring.

[0145] Among them, training a stirring state recognizer including X stirring state recognition branches in the cloud server includes:

[0146] According to the stirring data of the same food type in the historical time, collect the sample motor power sequence set, and collect the stirring state information of the food under different sample motor power sequences to obtain the sample stirring state information set;

[0147] Divide the sample motor power sequence set and the sample stirring state information set to obtain X stirring state recognition training data;

[0148] Respectively use the X stirring state recognition training data to train X stirring state recognition branches to obtain the stirring state recognizer.

[0149] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0150] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0151] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded computers, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0152] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0154] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concepts.

[0155] Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A switch control method for a household blender, characterized in that: The method comprises: The weight and type of the target food to be blended are obtained through a weight sensor in the blender, and the blending time is predicted to obtain the predicted blending time; During the starting stirring process, the motor power is monitored and recorded to obtain the motor power sequence; Processing and obtaining the time coefficient of the cumulative stirring time and the predicted stirring time, analyzing and obtaining the power variation coefficient of the motor power sequence, and configuring the stirring state identification coefficient; According to the stirring state identification coefficient, the stirring state is identified based on the motor power sequence and the accumulated stirring time to obtain stirring state information, and the stirrer is turned off when a preset stirring state is met.

2. The switch control method for a household blender according to claim 1, characterized in that: The weight and type of the target food to be blended are obtained through the weight sensor in the blender, including: Obtaining the weight of the target food to be blended through a weight sensor in the blender; Receive food type input from the user.

3. The switch control method for a household blender according to claim 2, characterized in that: Perform mixing time prediction and obtain predicted mixing time, including: According to the mixing history data of the same type of mixers, a sample food weight set and a sample food type set are collected; Collect the time required for stirring to reach the required stirring state under different sample food types and different sample food weights, and mark it as a sample stirring time set; Constructing a mapping relationship among the sample food weight set, the sample food type set and the sample mixing time set to obtain a mixing time prediction table; The food weight and food type are input into the mixing time prediction table, and the predicted mixing time is output.

4. The switch control method for a household blender according to claim 1, characterized in that: During the stirring process, the motor power is monitored and recorded to obtain the motor power sequence, including: During the stirring process, the motor power of the most recent multiple time stamps is continuously monitored; The motor powers of the most recent multiple timestamps are arranged in chronological order to obtain a motor power sequence.

5. The switch control method for a household blender according to claim 1, characterized in that: Processing and obtaining the time coefficient of the cumulative stirring time and the predicted stirring time includes: Monitor and obtain the cumulative mixing time after the mixer is started; According to the accumulated stirring time and the predicted stirring time, the time coefficient is obtained by processing, as shown in the following formula: Among them, KT is the time coefficient, T l is the cumulative stirring time, T y To predict the mixing time.

6. The switch control method for a household blender according to claim 1, characterized in that: Analyze and obtain the power variation coefficient of the motor power sequence and configure the stirring state identification coefficient, including: Randomly selecting two motor powers multiple times within the motor power sequence, and calculating the power fluctuation amplitude; Output the maximum power fluctuation amplitude as the power fluctuation coefficient; Subtract the power fluctuation coefficient from 1 to obtain the power variation coefficient; The stirring state recognition coefficient is calculated based on the time coefficient and the power variation coefficient.

7. The switch control method for a household blender according to claim 1, characterized in that: According to the stirring state recognition coefficient, the stirring state is recognized based on the motor power sequence and the accumulated stirring time, stirring state information is obtained, and when the preset stirring state is met, the stirrer is turned off, including: In the cloud server, a stirring state recognizer including X stirring state recognition branches is trained, where X is a positive integer; After the time coefficient is greater than or equal to the time coefficient threshold, Y is obtained by multiplying X by the stirring state identification coefficient and rounding; Randomly select Y stirring state recognition branches in the stirring state identifier, transmit the motor power sequence to the cloud server and input it, output and obtain the stirring state information of Y branches, and calculate and obtain the stirring state information; Determine whether the stirring state information satisfies the preset stirring state. If so, turn off the mixer. If not, determine whether the accumulated saturated stirring time with a time coefficient of 1 is greater than the saturated stirring time threshold. If so, turn off the mixer. If not, continue stirring.

8. The switch control method for a household blender according to claim 7, characterized in that: In the cloud server, a stirring state recognizer including X stirring state recognition branches is trained, including: According to the stirring data of the same type of food in the historical time, a set of sample motor power sequences is collected, and the stirring state information of the food under different sample motor power sequences is collected to obtain a set of sample stirring state information; Dividing the sample motor power sequence set and the sample stirring state information set to obtain X stirring state recognition training data; The X stirring state recognition training data are respectively used to train X stirring state recognition branches to obtain a stirring state recognizer.

9. A switch control system for a household blender, characterized in that: The system is used to execute the method according to any one of claims 1 to 8, and the system comprises: A mixing time prediction module is used to obtain the weight and type of the target food to be mixed through a weight sensor in the mixer, and to predict the mixing time to obtain the predicted mixing time; The motor power monitoring module is used to monitor and record the motor power during the starting of the stirring process and obtain the motor power sequence; A stirring identification configuration module is used to process and obtain the time coefficient of the cumulative stirring time and the predicted stirring time, analyze and obtain the power change coefficient of the motor power sequence, and configure the stirring state identification coefficient; The identification switch control module is used to identify the stirring state according to the stirring state identification coefficient, based on the motor power sequence and the accumulated stirring time, obtain stirring state information, and turn off the mixer when the preset stirring state is met.

Images