Method and system for testing energy efficiency of coffee machine

Abstract

The invention discloses a coffee machine energy efficiency test method and system, and the method comprises the steps: obtaining the working data and state data of a coffee machine, extracting the input power of the coffee machine according to a current signal, determining the actual energy consumption and basic heat efficiency of coffee making according to monitoring data, constructing a user influence degree and an environment influence factor according to the state data, and testing the energy efficiency of the coffee machine. And determining grinding efficiency, first heat efficiency and second heat efficiency, calculating a data correlation degree, and constructing a coffee machine energy efficiency test model according to the second heat efficiency, the correlation degree, the environmental impact factor, the grinding efficiency and the current signal, and inputting the working data and the state data of the coffee machine to be tested into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency index. The method not only can improve the precision and efficiency of the energy efficiency test of the coffee machine, but also has better interpretability, and can be directly applied to the energy efficiency test system of the coffee machine.

Description

Technical Field

[0001] The present invention relates to the field of energy efficiency measurement, and in particular to a coffee machine energy efficiency testing method and system. Background Art

[0002] With the intensification of the global energy crisis and the improvement of environmental awareness, energy efficiency testing and evaluation has become an important research direction in the field of household appliances. The popularity of coffee culture has made coffee machines an indispensable electrical appliance in many homes and offices. However, there are significant differences in the energy efficiency of different coffee machines. High-efficiency coffee machines can not only reduce electricity bills, but are also of great significance for reducing energy consumption and carbon emissions.

[0003] Traditional coffee machine energy efficiency evaluation methods are often based on simple power measurement, ignoring the complexity and diversity of influencing factors such as user operating habits, environmental factors and actual working degree during the actual working process of the coffee machine, making it difficult to fully reflect the energy efficiency of the coffee machine during actual use. In addition, when considering the energy efficiency test of the above factors, the data processing process is cumbersome, the data volume is large and the data types are many, which makes the efficiency and accuracy of the energy efficiency test low. With the rapid development of computer technology, based on big data technology to collect, process and analyze the working data and status data of the coffee machine during actual use, combined with machine learning technology to learn the data rules, it is possible to predict the energy efficiency of the coffee machine under specific working and environmental conditions. Therefore, by fully considering the actual working and environmental status of the coffee machine, combining big data and machine learning technology, an accurate and efficient coffee machine energy efficiency test method and system are designed to overcome the shortcomings of the traditional coffee machine energy efficiency evaluation method and system, which can provide a scientific basis for the design optimization and energy efficiency improvement of the coffee machine, and help the sustainable development strategic goal of global energy conservation and emission reduction. Summary of the invention

[0004] The object of the present invention is to provide a method and system for testing the energy efficiency of a coffee machine.

[0005] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0006] The present invention comprises the following steps:

[0007] Acquire working data and status data of the coffee machine, and pre-process the working data and the status data; the working data includes monitoring data and current signals;

[0008] Extracting the coffee machine input power according to the current signal, determining the actual energy consumption of coffee making according to the monitoring data, and determining the basic thermal efficiency according to the coffee machine input power and the actual energy consumption of coffee making; the coffee machine input power includes coffee making input power, coffee bean grinding power and standby power;

[0009] The state data is divided into user state data and environment state data, a user influence is constructed according to the user state data, a grinding efficiency is determined according to the user influence and the coffee bean grinding power, and a first thermal efficiency is determined according to the grinding efficiency, the user influence and the basic thermal efficiency;

[0010] Inputting the environmental state data into an environmental impact function to obtain an environmental impact factor, determining a second thermal efficiency according to the environmental impact factor and the first thermal efficiency, and respectively calculating the correlation between the second thermal efficiency and the grinding efficiency, the user impact degree, and the basic thermal efficiency;

[0011] A coffee machine energy efficiency test model is constructed according to the second thermal efficiency, the correlation degree, the environmental impact factor, the grinding efficiency and the current signal, and the working data and status data of the coffee machine to be tested are input into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency index.

[0012] Furthermore, the method for extracting the input power of the coffee machine according to the current signal includes:

[0013] The current signal of the original time series is subjected to the first empirical mode decomposition according to the preset modulus K to obtain K modal signals IMF1 and a residual signal R(t), which is expressed as:

[0014]

[0015] Where I(t) is the current signal at time t, a 1i (t) is the i-th IMF1 obtained by the first empirical mode decomposition of the current signal I(t) at time t;

[0016] Take the absolute value of IMF1 and construct the envelope function based on the extreme points using spline interpolation right The second empirical mode decomposition is performed to obtain N modal signals IMF2 and a residual signal R2(t), which is expressed as:

[0017]

[0018] where a 2ij (t) is the jth IMF2 obtained by the second empirical mode decomposition of the i-th IMF1 of the current signal I(t) at time t, R 2i (t) is the residual signal obtained by the second empirical mode decomposition of the i-th IMF1 at time t;

[0019] Calculate the mutual information MI of all IMF2 signal components and the original current signal I(t) ij , filter the effective signal components according to the mutual information size, and define the average mutual information as MI m, M.I. ij >MI m The corresponding IMF2 is valid information. According to the screening result, the real signal is Hilbert transformed to obtain the complex analytical signal I(t) expression:

[0020]

[0021] Where j is the imaginary unit, the Hilbert transform P is the Cauchy principal value at the transformation, and τ is the integral variable;

[0022] Calculate the instantaneous amplitude a(t) and instantaneous phase θ(t):

[0023]

[0024] The instantaneous power, i.e., the input power of the coffee machine, is obtained by calculating the square of the instantaneous amplitude, and the coffee making input power, coffee bean grinding power, and standby power are distinguished according to the time series; the coffee making input power includes heating power, extraction power, and whipping power.

[0025] Furthermore, the method for determining the basic thermal efficiency includes:

[0026] The actual energy consumption of coffee making is determined based on the monitoring data, and the expression is:

[0027]

[0028] Where Q is the actual energy consumption for making n cups of coffee, c1 is the specific heat capacity of water, V i is the volume change of water in the water tank when making the i-th cup of coffee, ρ is the density of water, ΔT 1i is the difference between the water temperature in the water tank and the preset heating temperature when making the i-th cup of coffee, m wi is the mass of the extract when making the i-th cup of coffee, ΔT 2i is the temperature difference of the extract before and after extraction when making the i-th cup of coffee, c2 is the specific heat capacity of milk, m mi is the mass of milk when making the i-th cup of coffee, ΔT 3i is the difference between the temperature of the milk in the milk foam cup and the temperature of the milk foam after being whipped when making the i-th cup of coffee;

[0029] According to the input power of coffee making and the actual energy consumption of coffee making, calculate the basic thermal efficiency of making n cups of coffee in a period of time:

[0030]

[0031] Where η0 is the thermal efficiency of making n cups of coffee in a period of time, w1 is the thermal efficiency weight when heating water, w2 is the thermal efficiency weight when extracting, and w3 is the thermal efficiency weight when frothing milk. When making the i-th cup of coffee, P 1i is the heating power, P 2i is the extraction power, P 3i is the power dissipation, t 1i is the heating time, t 2i is the extraction time, t 3i To pass the time, when making coffee without milk foam, only the comprehensive weight of thermal efficiency of heating and extraction stages is calculated.

[0032] Further, the method for determining the first thermal efficiency includes:

[0033] A clustering algorithm is used to divide the state data into user state data and environment state data;

[0034] The user influence of the coffee machine during a period of working time is constructed based on the user status data. The expression is:

[0035]

[0036] Influence1 is the user influence of the coffee machine during a period of working time, v1, v2, v3, v4 are the user influence weights, β is the working state coefficient, t F For continuous working time, is the average power during continuous working time, m is the number of effective rest time, t ri is the i-th effective rest time, U max is the maximum voltage during continuous working time, U min is the minimum voltage during continuous working time, N0 is the number of standard extraction cups, N i is the actual number of extraction cups during continuous working time, is the average power of each extraction during continuous working time, P2 is the standard extraction power, is the average duration of each extraction during continuous working hours, and t2 is the standard extraction duration;

[0037] The grinding efficiency is determined according to the user influence and the coffee bean grinding power, and the first thermal efficiency is determined according to the grinding efficiency, the user influence and the basic thermal efficiency. The expression is:

[0038]

[0039] Where E grinding is the grinding efficiency, η1 is the first thermal efficiency, n is the number of coffee cups made by the coffee machine within a certain working time, P4 is the standard grinding power, t4 is the standard grinding time, P 4iis the grinding power of the i-th cup, t 4i is the grinding time of the i-th cup, and ε is the first thermal efficiency coefficient.

[0040] Further, the method for determining the second thermal efficiency includes:

[0041] The environmental status data is input into the environmental impact function to obtain the environmental impact factor, which is expressed as:

[0042]

[0043] Influence2 is the environmental impact factor of the coffee machine, μ1, μ2, μ3, and μ4 are the weights of the environmental impact factors, and T 1i is the current ambient temperature, T aim is the preset heating temperature, γ heating influence coefficient, T 2i is the current machine operating temperature, T0 is the standard machine operating temperature, H i is the current ambient humidity, H0 is the standard ambient humidity, t0 is the heating time of a unit of water under standard pressure, ρ is the density of water, g is the acceleration of gravity, h is the current altitude, Pa i is the current atmospheric pressure, Pa0 is the standard atmospheric pressure, and δ is a bias constant to prevent the denominator from being zero;

[0044] The second thermal efficiency is determined according to the environmental impact factor and the first thermal efficiency, and the expression is:

[0045]

[0046] Where η2 is the second thermal efficiency, Therefore, the second thermal efficiency is less than the first thermal efficiency. is the second thermal efficiency influence coefficient.

[0047] Further, the method for calculating the correlation between the second thermal efficiency and the grinding efficiency, the user influence and the basic thermal efficiency includes:

[0048] Arrange the second thermal efficiency, grinding efficiency, user influence and basic thermal efficiency according to the time series, and set the second thermal efficiency sequence X = {x1, x2, ..., x n} is the reference sequence, n is the sequence length, x i For the i-th observation value of the sequence X, define Y c ={y c1 ,y c2 ,…,y cn} is the observation sequence, y ci For sequence Y cThe i-th observation value in the above sequence, c∈{1,2,3} represents the observation category, Y1 is the basic thermal efficiency sequence, Y2 is the grinding efficiency sequence, and Y3 is the user influence sequence;

[0049] According to the reference sequence X and the observed sequence Y c The distance parameter of the sequence is constructed by the average distance of:

[0050]

[0051] in is the reference sequence X and the observation sequence Y c The distance parameter, is the reference sequence X and the observation sequence Y c The average distance is the average distance between the reference sequence X and the x-axis;

[0052] Calculate the reference sequence X and the observed sequence Y c The comprehensive correlation degree is expressed as:

[0053]

[0054] where r(X,Y c ) is the reference sequence X and the observed sequence Y c The comprehensive correlation of is the association weight, and ρ∈[0,1] is the resolution coefficient.

[0055] Furthermore, the method for obtaining the energy efficiency index of the coffee machine includes:

[0056] The second thermal efficiency, correlation, environmental impact factor, grinding efficiency and current signal are combined into a comprehensive data set, and the comprehensive data set is divided into a training set and a test set at a ratio of 7:3;

[0057] Construct a coffee machine energy efficiency test model, including random forest algorithm, genetic algorithm and BP neural network;

[0058] The training set data is preprocessed through the input layer of the coffee machine energy efficiency test model;

[0059] The random forest algorithm evaluates the importance of each feature in the input data, reduces the input dimension, and provides an important feature subset for the BP neural network;

[0060] The genetic algorithm finds the optimal parameter setting based on the feature subset provided by the random forest algorithm and initializes the initial weights and thresholds of the BP neural network;

[0061] The BP neural network learns according to the feature subset provided by the random forest algorithm and outputs the predicted value of the energy efficiency index; the hidden layer of the BP neural network uses the tansig activation function, the output layer uses the purelin activation function, and the training algorithm uses the scaled conjugate gradient algorithm;

[0062] The mean square error is used to measure the difference between the predicted value and the true value, the holdout method is used to verify the generalization ability of the model, the parameters of the random forest algorithm and the genetic algorithm are adjusted according to the evaluation results, the AdaDelta algorithm is used to optimize the performance of the BP neural network prediction model, and the test set is used to evaluate the coffee machine energy efficiency test model;

[0063] The current signal and status data of the coffee machine to be tested are input into the coffee machine energy efficiency test model to obtain three types of coffee machine energy efficiency indicators: thermal efficiency, grinding power and standby power.

[0064] In a second aspect, a coffee machine energy efficiency testing system includes:

[0065] Data acquisition module: including a thermometer, a liquid flow meter, a current sensor and a timer; used to collect the working data and status data of the coffee machine, and pre-process the working data and the status data;

[0066] Data processing module: used to extract the input power of the coffee machine, to calculate and determine the actual energy consumption of coffee making, to determine the basic thermal efficiency, to calculate the user influence, grinding efficiency and environmental influence factors, to determine the first thermal efficiency and the second thermal efficiency, and to calculate the correlation;

[0067] Model prediction module: used for constructing a coffee machine energy efficiency test model according to the second thermal efficiency, the correlation degree, the environmental impact factor, the grinding efficiency and the current signal, and inputting the working data and status data of the coffee machine to be tested into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency index;

[0068] Supervision module: used to store, view and manage the working data, the status data and the energy efficiency index of the coffee machine, and provide feedback to the user according to the energy efficiency index threshold of the coffee machine.

[0069] The beneficial effects of the present invention are:

[0070] The present invention is a coffee machine energy efficiency testing method and system. Compared with the prior art, the present invention has the following technical effects:

[0071] The present invention can improve the accuracy of the energy efficiency test of the coffee machine through modal decomposition and data transformation, determination of the actual energy consumption of coffee making, construction of user influence and environmental influence factors, data correction, calculation of correlation, and model construction steps, thereby improving the efficiency and precision of the energy efficiency test of the coffee machine. The energy efficiency test of the coffee machine is made intelligent, which can greatly save resources and improve the efficiency of the energy efficiency test. It can realize high-precision energy efficiency test of the coffee machine to be tested, and quickly perform high-precision energy efficiency test on the coffee machine to be tested, providing a scientific basis for the design optimization and energy efficiency improvement of the coffee machine, which is of great significance to reducing energy consumption and carbon emissions, and can adapt to different coffee machine energy efficiency test systems and the coffee machine energy efficiency test requirements of different users, and has a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 The present invention is a flow chart of the steps of a method for testing energy efficiency of a coffee machine. DETAILED DESCRIPTION

[0073] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0074] A coffee machine energy efficiency testing method and system of the present invention comprises the following steps:

[0075] like Figure 1 As shown, in this embodiment, the following steps are included:

[0076] Acquire working data and status data of the coffee machine, and pre-process the working data and the status data; the working data includes monitoring data and current signals;

[0077] Extracting the coffee machine input power according to the current signal, determining the actual energy consumption of coffee making according to the monitoring data, and determining the basic thermal efficiency according to the coffee machine input power and the actual energy consumption of coffee making; the coffee machine input power includes coffee making input power, coffee bean grinding power and standby power;

[0078] The state data is divided into user state data and environment state data, a user influence is constructed according to the user state data, a grinding efficiency is determined according to the user influence and the coffee bean grinding power, and a first thermal efficiency is determined according to the grinding efficiency, the user influence and the basic thermal efficiency;

[0079] Inputting the environmental state data into an environmental impact function to obtain an environmental impact factor, determining a second thermal efficiency according to the environmental impact factor and the first thermal efficiency, and respectively calculating the correlation between the second thermal efficiency and the grinding efficiency, the user impact degree, and the basic thermal efficiency;

[0080] A coffee machine energy efficiency test model is constructed according to the second thermal efficiency, the correlation degree, the environmental impact factor, the grinding efficiency and the current signal, and the working data and status data of the coffee machine to be tested are input into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency index.

[0081] In this embodiment, the method for extracting the input power of the coffee machine according to the current signal includes:

[0082] The current signal of the original time series is subjected to the first empirical mode decomposition according to the preset modulus K to obtain K modal signals IMF1 and a residual signal R(t), which is expressed as:

[0083]

[0084] Where I(t) is the current signal at time t, a 1i (t) is the i-th IMF1 obtained by the first empirical mode decomposition of the current signal I(t) at time t;

[0085] Take the absolute value of IMF1 and construct the envelope function based on the extreme points using spline interpolation right The second empirical mode decomposition is performed to obtain N modal signals IMF2 and a residual signal R2(t), which is expressed as:

[0086]

[0087] where a 2ij (t) is the jth IMF2 obtained by the second empirical mode decomposition of the i-th IMF1 of the current signal I(t) at time t, R 2i (t) is the residual signal obtained by the second empirical mode decomposition of the i-th IMF1 at time t;

[0088] Calculate the mutual information MI of all IMF2 signal components and the original current signal I(t) ij , filter the effective signal components according to the mutual information size, and define the average mutual information as MI m , M.I. ij >MI m The corresponding IMF2 is valid information. According to the screening result, the real signal is Hilbert transformed to obtain the complex analytical signal I(t) expression:

[0089]

[0090] Where j is the imaginary unit, the Hilbert transform P is the Cauchy principal value at the transformation, and τ is the integral variable;

[0091] Calculate the instantaneous amplitude a(t) and instantaneous phase θ(t):

[0092]

[0093] The instantaneous power, i.e., the input power of the coffee machine, is obtained by calculating the square of the instantaneous amplitude, and the coffee making input power, coffee bean grinding power, and standby power are distinguished according to the time series; the coffee making input power includes heating power, extraction power, and whipping power;

[0094] In the actual evaluation, XX brand fully automatic coffee machine was used to make three types of coffee according to the standards of full-brew American, ordinary American and ordinary latte. 5 cups of each type of coffee were made, and the current information of the three types of coffee during production in different time periods was obtained. The machine power at different coffee production stages was calculated according to the above method, and the average value was obtained: 1. Full-brew American (heating power 724.8W, extraction power 360W); 2. Ordinary American (heating power 920W, extraction power 315W); 3. Ordinary latte American (heating power 915W, extraction power 295W, whipping power 930W), and the networked standby power was 1W.

[0095] In this embodiment, the method for determining the basic thermal efficiency includes:

[0096] The monitoring data includes water flow, water temperature change and time data of all stages of coffee extraction; the data acquisition method includes setting a flow meter and a thermometer at the water outlet of the water tank, setting a thermometer at the liquid inlet and outlet of the extraction device, weighing the mass of the extraction liquid, setting a thermometer in the milk froth cup, and determining the working time of all stages through the change of the current signal;

[0097] The actual energy consumption of coffee making is determined based on the monitoring data, and the expression is:

[0098]

[0099] Where Q is the actual energy consumption for making n cups of coffee, c1 is the specific heat capacity of water, V i is the volume change of water in the water tank when making the i-th cup of coffee, ρ is the density of water, ΔT 1i is the difference between the water temperature in the water tank and the preset heating temperature when making the i-th cup of coffee, m wi is the mass of the extract when making the i-th cup of coffee, ΔT 2i is the temperature difference of the extract before and after extraction when making the i-th cup of coffee, c2 is the specific heat capacity of milk, m mi is the mass of milk when making the i-th cup of coffee, ΔT 3i is the difference between the temperature of the milk in the milk foam cup and the temperature of the milk foam after being whipped when making the i-th cup of coffee;

[0100] According to the input power of coffee making and the actual energy consumption of coffee making, calculate the basic thermal efficiency of making n cups of coffee in a period of time:

[0101]

[0102] Where η0 is the thermal efficiency of making n cups of coffee in a period of time, w1 is the thermal efficiency weight when heating water, w2 is the thermal efficiency weight when extracting, and w3 is the thermal efficiency weight when frothing milk. When making the i-th cup of coffee, P 1i is the heating power, P 2i is the extraction power, P 3i is the power dissipation, t 1i is the heating time, t 2i is the extraction time, t 3i To pass the time, when making coffee without milk foam, only the comprehensive weight of thermal efficiency in the heating and extraction stages is calculated;

[0103] In the actual evaluation, XX brand fully automatic coffee machine was used to make three types of coffee according to the standards of full-brew Americano, regular Americano and regular latte. Five cups of each type of coffee were made, and the average monitoring data of the three types of coffee during different time periods were obtained: 1. Full-brew Americano (water heating time 45s, initial water temperature 23°C, heating water temperature 80.5°C, water flow volume 120ml, extraction liquid temperature 95°C, extraction liquid mass 63g, extraction time 25s); 2. Regular Americano (water heating time 3. Ordinary American Latte (water heating time 30s, initial water temperature 20℃, heating water temperature 89℃, water flow volume 100ml, extraction liquid temperature 83℃, extraction liquid mass 41g, extraction time 15s, milk mass 155g, whipping time 35s, pre-whipping temperature 24℃, post-whipping temperature 50℃);

[0104] When making American coffee, w1=0.1, w2=0.9; when making latte, w1=0.05, w2=0.6, w3=0.35. The basic thermal efficiency of the coffee machine when making the three types of coffee is: 45.962% for full-brew American coffee, 49.28% for regular American coffee and 43.59% for regular latte.

[0105] In this embodiment, the method for determining the first thermal efficiency includes:

[0106] A clustering algorithm is used to divide the state data into user state data and environment state data;

[0107] The user influence of the coffee machine during a period of working time is constructed based on the user status data. The expression is:

[0108]

[0109] Influence1 is the user influence of the coffee machine during a period of working time, v1, v2, v3, v4 are the user influence weights, β is the working state coefficient, t F For continuous working time, is the average power during continuous working time, m is the number of effective rest time, t ri is the i-th effective rest time, U max is the maximum voltage during continuous working time, U min is the minimum voltage during continuous working time, N0 is the number of standard extraction cups, N i is the actual number of extraction cups during continuous working time, is the average power of each extraction during continuous working time, P2 is the standard extraction power, is the average duration of each extraction during continuous working hours, and t2 is the standard extraction duration;

[0110] The grinding efficiency is determined according to the user influence and the coffee bean grinding power, and the first thermal efficiency is determined according to the grinding efficiency, the user influence and the basic thermal efficiency. The expression is:

[0111]

[0112] Where E grinding is the grinding efficiency, η1 is the first thermal efficiency, n is the number of coffee cups made by the coffee machine within a certain working time, P4 is the standard grinding power, t4 is the standard grinding time, P 4i is the grinding power of the i-th cup, t 4i is the grinding time of the i-th cup, ε is the first thermal efficiency coefficient;

[0113] In the actual evaluation, the continuous working time of the coffee machine is defined as no rest time exceeding 20 minutes in the middle, and the effective rest time is defined as the rest time between two coffee making of the same coffee machine, which is between 5 and 20 minutes. When making full-brew Americano, ordinary Americano and ordinary latte, the coffee machine works continuously for 3 hours, 1.5 hours and 0 hours in each state. The standard production speed of the coffee machine is 10 cups / h, the standard extraction power is 300w, and the standard extraction time is 15s. The user influence of the coffee machine when making full-brew Americano, ordinary Americano and ordinary latte is calculated from the monitoring data to be 1.45, 1.17 and 0.95 respectively;

[0114] Through the current data, it is obtained that the average grinding power of the coffee machine when making full-brew American, regular American and regular latte is 710W, 705W and 600w respectively, ε=0.35, and the grinding efficiency and first thermal efficiency of making the three types of coffee are calculated as follows: 1. Full-brew American: grinding efficiency 1.045, first thermal efficiency 41.68%; 2. Regular American: grinding efficiency 1.128, first thermal efficiency 48.85%; 1. Full-brew American: grinding efficiency 1.224, first thermal efficiency 42.11%.

[0115] In this embodiment, the method for determining the second thermal efficiency includes:

[0116] The environmental status data is input into the environmental impact function to obtain the environmental impact factor, which is expressed as:

[0117]

[0118] Influence2 is the environmental impact factor of the coffee machine, μ1, μ2, μ3, and μ4 are the weights of the environmental impact factors, and T 1i is the current ambient temperature, T aim is the preset heating temperature, γ heating influence coefficient, T 2i is the current machine operating temperature, T0 is the standard machine operating temperature, H i is the current ambient humidity, H0 is the standard ambient humidity, t0 is the heating time of a unit of water under standard pressure, ρ is the density of water, g is the acceleration of gravity, h is the current altitude, Pa i is the current atmospheric pressure, Pa0 is the standard atmospheric pressure, and δ is a bias constant to prevent the denominator from being zero;

[0119] The second thermal efficiency is determined according to the environmental impact factor and the first thermal efficiency, and the expression is:

[0120]

[0121] Where η2 is the second thermal efficiency, Therefore, the second thermal efficiency is less than the first thermal efficiency. is the second thermal efficiency influence coefficient;

[0122] In the actual evaluation, the average monitoring data of the three types of coffee made in different time periods were obtained: 1. Full-brew American (current machine operating temperature 92°C, current ambient humidity 45%, current atmospheric pressure 98kPa); 2. Ordinary American (current machine operating temperature 95°C, current ambient humidity 60%, current atmospheric pressure 100kPa); 3. Ordinary Latte American (current machine operating temperature 90°C, current ambient humidity 40%, current atmospheric pressure 101kPa), the preset heating temperature is 100°C, the standard machine operating temperature is 95°C, the standard ambient humidity is 55%, and the standard atmospheric pressure is 98kPa. According to the monitoring data, it was calculated that: 1. Full-brew American: environmental impact factor 6.6, second thermal efficiency 38.86%; 2. Ordinary American: environmental impact factor 7.4, second thermal efficiency 47.91%; 1. Full-brew American: environmental impact factor 7.2, second thermal efficiency 40.79%.

[0123] In this embodiment, the method for calculating the correlation between the second thermal efficiency and the grinding efficiency, the user influence and the basic thermal efficiency includes:

[0124] Arrange the second thermal efficiency, grinding efficiency, user influence and basic thermal efficiency according to the time series, and set the second thermal efficiency sequence X = {x1, x2, ..., x n} is the reference sequence, n is the sequence length, x i For the i-th observation value of the sequence X, define Y c ={y c1 ,y c2 ,…,y cn} is the observation sequence, y ci For sequence Y c The i-th observation value in the above sequence, c∈{1,2,3} represents the observation category, Y1 is the basic thermal efficiency sequence, Y2 is the grinding efficiency sequence, and Y3 is the user influence sequence;

[0125] According to the reference sequence X and the observed sequence Y c The distance parameter of the sequence is constructed by the average distance of:

[0126]

[0127] in is the reference sequence X and the observation sequence Y c The distance parameter, is the reference sequence X and the observation sequence Y c The average distance is the average distance between the reference sequence X and the x-axis;

[0128] Calculate the reference sequence X and the observed sequence Y c The comprehensive correlation degree is expressed as:

[0129]

[0130] where r(X,Y c ) is the reference sequence X and the observed sequence Y c The comprehensive correlation of is the association weight, ρ∈[0,1] is the resolution coefficient;

[0131] In this embodiment, the method for obtaining the energy efficiency index of the coffee machine includes:

[0132] The second thermal efficiency, correlation, environmental impact factor, grinding efficiency and current signal are combined into a comprehensive data set, and the comprehensive data set is divided into a training set and a test set at a ratio of 7:3;

[0133] Construct a coffee machine energy efficiency test model, including random forest algorithm, genetic algorithm and BP neural network;

[0134] The training set data is preprocessed through the input layer of the coffee machine energy efficiency test model;

[0135] The random forest algorithm evaluates the importance of each feature in the input data, reduces the input dimension, and provides an important feature subset for the BP neural network;

[0136] The genetic algorithm finds the optimal parameter setting based on the feature subset provided by the random forest algorithm and initializes the initial weights and thresholds of the BP neural network;

[0137] The BP neural network learns according to the feature subset provided by the random forest algorithm and outputs the predicted value of the energy efficiency index; the hidden layer of the BP neural network uses the tansig activation function, the output layer uses the purelin activation function, and the training algorithm uses the scaled conjugate gradient algorithm;

[0138] The mean square error is used to measure the difference between the predicted value and the true value, the holdout method is used to verify the generalization ability of the model, the parameters of the random forest algorithm and the genetic algorithm are adjusted according to the evaluation results, the AdaDelta algorithm is used to optimize the performance of the BP neural network prediction model, and the test set is used to evaluate the coffee machine energy efficiency test model;

[0139] The current signal and state data of the coffee machine to be tested are input into the coffee machine energy efficiency test model to obtain three types of coffee machine energy efficiency indicators: thermal efficiency, grinding power and standby power;

[0140] In the actual evaluation, multiple groups of data monitoring data numbers and status data are collected to train the test model, and the current signals and status data of the coffee machines to be tested in this group are input into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency indicators: 1. Full-extracted American: thermal efficiency 39.80%, grinding power 710W, standby power 1W; 2. Ordinary American: thermal efficiency 48.93%, grinding power 705W, standby power 1W; 3. Ordinary latte: thermal efficiency 41.79%, grinding power 600W, standby power 1W. According to the first-level energy efficiency assessment standard of fully automatic press machines, the thermal efficiency is ≥40%, and the network power is ≤1.5w. This pressure coffee machine basically meets the first-level energy efficiency standard, but the energy efficiency is low when making full-extracted American coffee.

[0141] In a second aspect, a coffee machine energy efficiency testing system includes:

[0142] Data acquisition module: including a thermometer, a liquid flow meter, a current sensor and a timer; used to collect the working data and status data of the coffee machine, and pre-process the working data and the status data;

[0143] Data processing module: used to extract the input power of the coffee machine, to calculate and determine the actual energy consumption of coffee making, to determine the basic thermal efficiency, to calculate the user influence, grinding efficiency and environmental influence factors, to determine the first thermal efficiency and the second thermal efficiency, and to calculate the correlation;

[0144] Model prediction module: used for constructing a coffee machine energy efficiency test model according to the second thermal efficiency, the correlation degree, the environmental impact factor, the grinding efficiency and the current signal, and inputting the working data and status data of the coffee machine to be tested into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency index;

[0145] Supervision module: used to store, view and manage the working data, the status data and the energy efficiency index of the coffee machine, and provide feedback to the user according to the energy efficiency index threshold of the coffee machine.

[0146] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A coffee machine energy efficiency testing method, characterized in that: The following steps are involved: S1. Acquire working data and status data of the coffee machine, and pre-process the working data and status data; the working data includes monitoring data and current signals; S2, extracting the coffee machine input power according to the current signal, determining the actual energy consumption of coffee making according to the monitoring data, and determining the basic thermal efficiency according to the coffee machine input power and the actual energy consumption of coffee making; the coffee machine input power includes coffee making input power, coffee bean grinding power and standby power; S3, dividing the state data into user state data and environment state data, constructing a user influence according to the user state data, determining a grinding efficiency according to the user influence and the coffee bean grinding power, and determining a first thermal efficiency according to the grinding efficiency, the user influence and the basic thermal efficiency; S4, inputting the environmental state data into an environmental impact function to obtain an environmental impact factor, determining a second thermal efficiency according to the environmental impact factor and the first thermal efficiency, and respectively calculating the correlation between the second thermal efficiency and the grinding efficiency, the user impact degree, and the basic thermal efficiency; S5. Construct a coffee machine energy efficiency test model according to the second thermal efficiency, the correlation degree, the environmental impact factor, the grinding efficiency and the current signal, and input the working data and status data of the coffee machine to be tested into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency index.

2. A coffee machine energy efficiency testing method according to claim 1, characterized in that: The method for extracting the input power of a coffee machine according to the current signal comprises: The current signal of the original time series is subjected to the first empirical mode decomposition according to the preset modulus K to obtain K modal signals IMF1 and a residual signal R(t), which is expressed as: Where I(t) is the current signal at time t, a 1i (t) is the i-th IMF1 obtained by the first empirical mode decomposition of the current signal I(t) at time t; Take the absolute value of IMF1 and construct the envelope function based on the extreme points using spline interpolation right The second empirical mode decomposition is performed to obtain N modal signals IMF2 and a residual signal R2(t), which is expressed as: where a 2ij (t) is the jth IMF2 obtained by the second empirical mode decomposition of the i-th IMF1 of the current signal I(t) at time t, R 2i (t) is the residual signal obtained by the second empirical mode decomposition of the i-th IMF1 at time t; Calculate the mutual information MI of all IMF2 signal components and the original current signal I(t) ij , filter the effective signal components according to the mutual information size, and define the average mutual information as MI m , M.I. ij >MI m The corresponding IMF2 is valid information. According to the screening result, the real signal is Hilbert transformed to obtain the complex analytical signal I(t) expression: Where j is the imaginary unit, the Hilbert transform P is the Cauchy principal value at the transformation, and τ is the integral variable; Calculate the instantaneous amplitude a(t) and instantaneous phase θ(t): The instantaneous power, i.e., the input power of the coffee machine, is obtained by calculating the square of the instantaneous amplitude, and the coffee making input power, coffee bean grinding power, and standby power are distinguished according to the time series; the coffee making input power includes heating power, extraction power, and whipping power.

3. A coffee machine energy efficiency testing method according to claim 1, characterized in that: The method for determining the basic thermal efficiency comprises: The actual energy consumption of coffee making is determined based on the monitoring data, and the expression is: Where Q is the actual energy consumption for making n cups of coffee, c1 is the specific heat capacity of water, V i is the volume change of water in the water tank when making the i-th cup of coffee, ρ is the density of water, ΔT 1i is the difference between the water temperature in the water tank and the preset heating temperature when making the i-th cup of coffee, m wi is the mass of the extract when making the i-th cup of coffee, ΔT 2i is the temperature difference of the extract before and after extraction when making the i-th cup of coffee, c2 is the specific heat capacity of milk, m mi is the mass of milk when making the i-th cup of coffee, ΔT 3i is the difference between the temperature of the milk in the milk foam cup and the temperature of the milk foam after being whipped when making the i-th cup of coffee; According to the input power of coffee making and the actual energy consumption of coffee making, calculate the basic thermal efficiency of making n cups of coffee in a period of time: Where η0 is the thermal efficiency of making n cups of coffee in a period of time, w1 is the thermal efficiency weight when heating water, w2 is the thermal efficiency weight when extracting, and w3 is the thermal efficiency weight when frothing milk. When making the i-th cup of coffee, P 1i is the heating power, P 2i is the extraction power, P 3i is the power dissipation, t 1i is the heating time, t 2i is the extraction time, t 3i To pass the time, when making coffee without milk foam, only the comprehensive weight of thermal efficiency of heating and extraction stages is calculated.

4. A coffee machine energy efficiency testing method according to claim 1, characterized in that: The method for determining the first thermal efficiency comprises: A clustering algorithm is used to divide the state data into user state data and environment state data; The user influence of the coffee machine during a period of working time is constructed based on the user status data. The expression is: Influence1 is the user influence of the coffee machine during a period of working time, v1, v2, v3, v4 are the user influence weights, β is the working state coefficient, t F For continuous working time, is the average power during continuous working time, m is the number of effective rest time, t ri is the i-th effective rest time, U max is the maximum voltage during continuous working time, U min is the minimum voltage during continuous working time, N0 is the number of standard extraction cups, N i is the actual number of extraction cups during continuous working time, is the average power of each extraction during continuous working time, P2 is the standard extraction power, is the average duration of each extraction during continuous working hours, and t2 is the standard extraction duration; The grinding efficiency is determined according to the user influence and the coffee bean grinding power, and the first thermal efficiency is determined according to the grinding efficiency, the user influence and the basic thermal efficiency. The expression is: Where E grinding is the grinding efficiency, η1 is the first thermal efficiency, n is the number of coffee cups made by the coffee machine within a certain working time, P4 is the standard grinding power, t4 is the standard grinding time, P 4i is the grinding power of the i-th cup, t 4i is the grinding time of the i-th cup, and ε is the first thermal efficiency coefficient.

5. A coffee machine energy efficiency testing method according to claim 1, characterized in that: A method for determining the second thermal efficiency comprises: The environmental status data is input into the environmental impact function to obtain the environmental impact factor, which is expressed as: Influence2 is the environmental impact factor of the coffee machine, μ1, μ2, μ3, and μ4 are the weights of the environmental impact factors, and T 1i is the current ambient temperature, T aim is the preset heating temperature, γ heating influence coefficient, T 2i is the current machine operating temperature, T0 is the standard machine operating temperature, H i is the current ambient humidity, H0 is the standard ambient humidity, t0 is the heating time of a unit of water under standard pressure, ρ is the density of water, g is the acceleration of gravity, h is the current altitude, Pa i is the current atmospheric pressure, Pa0 is the standard atmospheric pressure, and δ is a bias constant to prevent the denominator from being zero; The second thermal efficiency is determined according to the environmental impact factor and the first thermal efficiency, and the expression is: Where η2 is the second thermal efficiency, Therefore, the second thermal efficiency is less than the first thermal efficiency, and θ is the second thermal efficiency influence coefficient.

6. A coffee machine energy efficiency testing method according to claim 1, characterized in that: The method for calculating the correlation between the second thermal efficiency and the grinding efficiency, the user influence and the basic thermal efficiency comprises: Arrange the second thermal efficiency, grinding efficiency, user influence and basic thermal efficiency according to the time series, and set the second thermal efficiency sequence X = {x1, x2, ..., x n } is the reference sequence, n is the sequence length, x i For the i-th observation value of the sequence X, define Y c ={y c1 ,y c2 ,…,y cn } is the observation sequence, y ci For sequence Y c The i-th observation value in the above sequence, c∈{1,2,3} represents the observation category, Y1 is the basic thermal efficiency sequence, Y2 is the grinding efficiency sequence, and Y3 is the user influence sequence; According to the reference sequence X and the observed sequence Y c The distance parameter of the sequence is constructed by the average distance of: in is the reference sequence X and the observation sequence Y c The distance parameter, is the reference sequence X and the observation sequence Y c The average distance is the average distance between the reference sequence X and the x-axis; Calculate the reference sequence X and the observed sequence Y c The comprehensive correlation degree is expressed as: where r(X,Y c ) is the reference sequence X and the observed sequence Y c The comprehensive correlation of is the association weight, and ρ∈[0,1] is the resolution coefficient.

7. A coffee machine energy efficiency testing method according to claim 1, characterized in that: The method for obtaining the energy efficiency index of the coffee machine comprises: The second thermal efficiency, correlation, environmental impact factor, grinding efficiency and current signal are combined into a comprehensive data set, and the comprehensive data set is divided into a training set and a test set at a ratio of 7:3; Construct a coffee machine energy efficiency test model, including random forest algorithm, genetic algorithm and BP neural network; The training set data is preprocessed through the input layer of the coffee machine energy efficiency test model; The random forest algorithm evaluates the importance of each feature in the input data, reduces the input dimension, and provides an important feature subset for the BP neural network; The genetic algorithm finds the optimal parameter setting based on the feature subset provided by the random forest algorithm and initializes the initial weights and thresholds of the BP neural network; The BP neural network learns according to the feature subset provided by the random forest algorithm and outputs the predicted value of the energy efficiency index; the hidden layer of the BP neural network uses the tansig activation function, the output layer uses the purelin activation function, and the training algorithm uses the scaled conjugate gradient algorithm; The mean square error is used to measure the difference between the predicted value and the true value, the holdout method is used to verify the generalization ability of the model, the parameters of the random forest algorithm and the genetic algorithm are adjusted according to the evaluation results, the AdaDelta algorithm is used to optimize the performance of the BP neural network prediction model, and the test set is used to evaluate the coffee machine energy efficiency test model; The current signal and status data of the coffee machine to be tested are input into the coffee machine energy efficiency test model to obtain three types of coffee machine energy efficiency indicators: thermal efficiency, grinding power and standby power.

8. A coffee machine energy efficiency testing system, used to execute the method according to any one of claims 1 to 7, characterized in that: include: Data acquisition module: including thermometer, liquid flow meter, current sensor and timer; Used to collect working data and status data of the coffee machine, and pre-process the working data and status data; Data processing module: used to extract the input power of the coffee machine, to calculate and determine the actual energy consumption of coffee making, to determine the basic thermal efficiency, to calculate the user influence, grinding efficiency and environmental influence factors, to determine the first thermal efficiency and the second thermal efficiency, and to calculate the correlation; Model prediction module: used for constructing a coffee machine energy efficiency test model according to the second thermal efficiency, the correlation degree, the environmental impact factor, the grinding efficiency and the current signal, and inputting the working data and status data of the coffee machine to be tested into the coffee machine energy efficiency test model to obtain the coffee machine energy efficiency index; Supervision module: used to store, view and manage the working data, the status data and the energy efficiency index of the coffee machine, and provide feedback to the user according to the energy efficiency index threshold of the coffee machine.

Images