An oil-frying production line energy-saving management system and method based on artificial intelligence

Abstract

The application discloses an energy-saving management system and method for a frying production line based on artificial intelligence, relates to the technical field of energy data management, collects characteristic data of temperature change of products in a processing process, collects the characteristic data to obtain reference samples of product temperature change, obtains change periods of power of each device, obtains reference functions of power change of each device, samples instantaneous power of each device in a certain frying production line to obtain a power sample value sequence, performs clustering analysis on a plurality of power sequences to obtain a reference power sequence of the certain frying production line, compares a reference function set and a reference power consumption sequence of a target production line, evaluates power differences of the target production line power and reference values, verifies whether production quality is normal, and provides a plurality of alarm information in a hierarchical classification alarm mode, so that the energy consumption status of the production line is accurately reflected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy data management, and particularly relates to an oil-frying production line energy-saving management system and method based on artificial intelligence. BACKGROUND

[0002] The oil-frying production line is a high-energy-consumption food production line. How to manage the energy consumption of the oil-frying production line while ensuring product quality is a hot issue in the relevant field. The Chinese invention with the publication number CN116168025A and the name of an oil curtain type oil-fried peanut production system discloses a food oil-frying system. The energy consumption of the traditional immersion frying method is reduced through multi-stage oil curtain spraying.

[0003] In the multi-stage oil curtain spraying oil-frying method, the traditional immersion frying method is divided into multiple stages, and energy data is generated in each stage. The energy consumption of each stage needs to be managed in detail, which puts forward new requirements for the design precision and working mode of the production line energy management system. Therefore, the traditional oil-frying production line energy management system cannot perform the management task of multi-stage energy data. SUMMARY

[0004] The present application relates to the technical field of energy data management, and particularly relates to an oil-frying production line energy-saving management system and method based on artificial intelligence.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme: an oil-frying production line energy-saving management method based on artificial intelligence, the method comprising:

[0006] Step S100: when the oil-frying production line is running in a reference state, collecting characteristic data of temperature change of products in the processing process, and collecting the characteristic data to obtain a reference sample of product temperature change;

[0007] Step S200: collecting historical power records of each device in the oil-frying production line, obtaining the change period of the power of each device, obtaining a reference function of the power change of each device, and collecting the reference functions to obtain a reference function set of the oil-frying production line;

[0008] Step S300: sampling the instantaneous power of each device in a certain oil-frying production line to obtain a power sample value sequence, performing cluster analysis on a plurality of power sample value sequences to obtain a reference power sequence of the certain oil-frying production line;

[0009] Step S400: calculating a power difference evaluation value for the certain oil-frying production line as a target production line;

[0010] Step S500: when the power difference is evaluated to exceed the first evaluation threshold, a first type of alarm prompt is performed, the temperature record of the product processed in the target production line in a unit period is collected, and the difference degree of the temperature record and the reference sample is calculated, when the difference degree exceeds the second evaluation threshold, a second type of alarm prompt is performed.

[0011] Further, step S100 comprises:

[0012] Step S101: dividing the processing area of the frying processing equipment in the frying production line into a plurality of unit areas;

[0013] Step S102: collecting the product surface temperature of the product after oiling in all unit areas in the production process, obtaining the average value and variance of the product temperature after the first oiling operation, and forming the reference data set of the first oiling operation by the average value and variance;

[0014] Step S103: collecting all reference data sets in the production process under a certain standard state, arranging the reference data sets according to the order of the oiling process, and obtaining the reference sample.

[0015] Further, step S200 comprises:

[0016] Step S201: taking a certain device in the frying production line as a reference device, sampling the power value of the reference device in a change period, collecting the sampling values of each sampling, and obtaining the power sampling sequence of the reference device;

[0017] Step S202: function fitting is performed on all sampling values in the power sampling sequence, and a function of the power change of the reference device with time is obtained, which is denoted as the power change function;

[0018] Step S203: collecting the power change functions of all devices in the frying production line to obtain the reference function set of the frying production line.

[0019] Further, step S300 comprises:

[0020] Step S301: obtaining a certain moment tr, obtaining the instantaneous power of all devices in a certain frying production line at tr, collecting to obtain the power sampling value sequence Qtr corresponding to tr, Qtr (p1, p2, p3, …, pn), wherein p1, p2, p3, … and pn represent the instantaneous power of the 1st, 2nd, 3rd, … and nth device in a certain frying production line at tr respectively;

[0021] Step S302: Obtain a power sample value sequence corresponding to each time of a certain frying production line, convert the power sample value sequence into a corresponding numerical vector, collect the corresponding vectors of all power sample value sequences into a sample sequence set, and obtain a clustering core of the sample sequence set through a clustering algorithm;

[0022] Step S303: Obtain the corresponding vector of the clustering core, and take the numerical sequence of the corresponding vector as a reference power sequence;

[0023] The power sample value sequence is converted into a vector in a high-dimensional space, the vector group is clustered, and the numerical characteristics of the vector group are extracted to represent the relationship between the power of each device in the same production line at the same time.

[0024] Further, step S400 includes:

[0025] Step S401: Take a certain device in the target production line as a target device, sample the power of the target device i times in a unit period, record a power value each time, and the time length of the unit period is less than or equal to the change period of the target device;

[0026] Step S402: Collect i power values to form a target power value set, record the jth power value in the target power value set as mj, obtain a reference function f of the target device, draw f in a coordinate system, obtain the coordinates of mj in the coordinate system, and calculate the shortest distance dj from mj to f;

[0027] Step S403: Obtain the shortest distance of each power value in the target power value set to the reference function f, and record the sum of all shortest distances as the power deviation value g of the target device;

[0028] Step S404: Collect the power deviation values of all devices in the target production line, calculate the average of the power deviation values, and obtain the power difference evaluation value W1 of the target production line.

[0029] Further, step S500 includes:

[0030] Step S501: Set a first evaluation threshold H1, when W1 is greater than H1, a first type of alarm is given to the management personnel of the target production line, further obtain all product temperature records after oiling operation of the frying processing device in the target production line in a unit period, and obtain the average value and variance of the product surface temperature after each oiling operation;

[0031] Step S502: Form a target data group corresponding to each oiling operation by taking the average value and variance of the product surface temperature after each oiling operation, collect all target data groups in the unit period to obtain a target data sequence;

[0032] In the temperature measurement process, the average temperature information and the dispersion degree of temperature need to be extracted, wherein the temperature needs to meet the design requirements of the production line, and the dispersion degree of temperature is as small as possible to ensure uniform temperature during the heating process;

[0033] Step S503: obtaining a reference sample of the frying processing equipment in the target production line, and corresponding the oiling process corresponding to each target data group in the target data sequence to the reference data group in the reference sample;

[0034] Step S504: obtaining the rth target data group (Ar, Br) in the target data sequence, wherein Ar represents the average value of the temperature in the rth target data group, Br represents the variance of the temperature in the rth target data group, obtaining the reference data group (Xr, Yr) corresponding to the rth target data group, and calculating the sample difference Ur of the rth target data group, Ur = a * | Ar - Xr | + b * (Br - Yr), wherein a and b are weight coefficients, and satisfy the conditions a > 0 and b > 0;

[0035] For temperature, the temperature after each oiling needs to meet the design temperature of the production line, and too high temperature is easy to cause energy waste and fried paste product, and too low temperature cannot complete the cooking task, and for the dispersion degree, the smaller the dispersion degree, the more uniform the product is heated;

[0036] Step S505: collecting the sample differences of all target data groups in the target data sequence, calculating the average value of the sample differences, and taking the average value as the difference degree evaluation value W2;

[0037] Step S506: setting a second evaluation threshold H2, and when W2 is greater than H2, the second type of alarm is given to the management personnel of the target production line.

[0038] In order to better realize the above method, an oil frying production line energy-saving management system based on artificial intelligence is also proposed, and the system comprises:

[0039] The reference sample management module, the reference function management module, the reference power sequence management module, the power difference evaluation module and the information reminding module, wherein the reference sample management module is used to collect the characteristic data of temperature change of the product in the processing process, and the reference sample of the temperature change of the product is obtained by collecting the characteristic data, the reference function management module is used to obtain the change period of the power of each device, and the reference function of the power change of each device is obtained, the reference function set of the oil frying production line is obtained by collecting the reference functions, the reference power sequence management module is used to perform cluster analysis on the power sequence to obtain the reference power sequence of a certain oil frying production line, the power difference evaluation module is used to calculate the power difference evaluation value of the target production line, and the information reminding module is used to manage the alarm condition and give information prompt when the alarm condition is met;

[0040] Further, the reference sample management module comprises a temperature monitoring unit, a reference data set management unit and a sample management unit, wherein the temperature monitoring unit is configured to manage the temperature of the product in the processing area, the reference data set management unit is configured to manage the reference data set, and the sample management unit is configured to collect all reference data sets in the standard state of the production process to obtain the reference sample.

[0041] Further, the reference function management module comprises a power sampling sequence management unit, a function fitting unit and a function set management unit, wherein the power sampling sequence management unit is configured to collect the sampling values to obtain the power sampling sequence of the reference device, the function fitting unit is configured to perform function fitting on the sampling values in the power sampling sequence to obtain the power change function, and the function set management unit is configured to collect the power change functions to manage the reference function set.

[0042] Further, the reference power sequence management module comprises a power sampling value sequence management unit, a clustering management unit and a reference sequence management unit, wherein the power sampling value sequence management unit is configured to collect the instantaneous power of all devices in the frying production line to manage the power sampling value sequence, the clustering management unit is configured to cluster the vectors corresponding to the power sampling values to obtain the clustering core, and the reference sequence management unit is configured to manage the reference power sequence.

[0043] Further, the power difference evaluation module comprises a power sampling unit, a distance management unit, a power deviation value management unit and a power difference evaluation value management unit, wherein the power sampling unit is configured to sample the power of the target device in a unit period, the distance management unit is configured to manage the shortest distance between each power value in the target power value set and the reference function, the power deviation value management unit is configured to calculate the power deviation value of the target device, and the power difference evaluation value management unit is configured to manage the power difference evaluation value of the target production line.

[0044] Further, the information reminding module comprises a first evaluation unit, a target data sequence management unit, a sample difference management unit, a second evaluation unit and an alarm feedback unit, wherein the first evaluation unit is configured to evaluate the numerical value of the power difference evaluation value, the target data sequence management unit is configured to obtain the target data sequence of the frying processing device in the target production line, the sample difference management unit is configured to correspond the target data sequence with the reference sample to calculate the sample difference, the second evaluation unit is configured to evaluate the sample difference, and the alarm feedback unit is configured to feed back the alarm information meeting the alarm condition to the relevant management personnel.

[0045] Compared with the prior art, the beneficial effects of the present application are that by collecting calibration data and historical operation records, a reference model of the power of the production line is established. The power and power consumption of the production equipment are compared from two angles by comparing each device in the production line with its own historical data and the factual data of other devices at the same period, so that the judgment of the power consumption of the device is more accurate. The production quality of the production line is further verified, and whether the production quality is normal is verified to obtain another reference value of the production of the device. A variety of alarm information is provided in a hierarchical classification alarm manner, and the energy consumption status of the production line is accurately reflected. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 It is a structural schematic diagram of an oil frying production line energy-saving management system based on artificial intelligence.

[0047] Figure 2 It is a flowchart of an oil frying production line energy-saving management method based on artificial intelligence. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0049] Embodiment: as shown in Figure 1 and Figure 2 The present application provides a technical solution, an oil frying production line energy-saving management method based on artificial intelligence:

[0050] Step S100: when the oil frying production line is running in a reference state, collecting characteristic data of temperature change of products in the processing process, and collecting the characteristic data to obtain a reference sample of temperature change of the products;

[0051] In the step S100, the step S100 includes:

[0052] Step S101: dividing the processing area of the oil frying processing equipment in the oil frying production line into a plurality of unit areas;

[0053] Step S102: collecting the surface temperature of the products after oiling in all unit areas in the production process to obtain the average value and variance of the product temperature after the first oiling operation, and the average value and variance form a reference data group of the first oiling operation;

[0054] Step S103: Collect all reference data sets in the production process under a certain standard state, arrange the reference data sets according to the order of the oiling process, and obtain a reference sample;

[0055] In the embodiment, an oil frying production line includes various devices, such as screening devices, drying devices, transportation devices, oil frying cooking devices, and cooling devices. The energy use and energy use efficiency of each device need to be monitored in the production process.

[0056] By collecting the historical production records of the production line, the appropriate production records are selected as the reference working state of the oil frying production line through the identification and screening of relevant experts.

[0057] Step S200: Collect the historical power records of each device in the oil frying production line, obtain the change period of the power of each device, obtain the reference function of the power change of each device, and collect the reference functions to obtain the reference function set of the oil frying production line.

[0058] The step S200 includes:

[0059] Step S201: Take a device in the oil frying production line as a reference device, sample the power value of the reference device in a change period, collect the sampling values of each sampling, and obtain a power sampling sequence of the reference device.

[0060] Step S202: Perform function fitting on all sampling values in the power sampling sequence to obtain a function of the power change of the reference device with time. The function is denoted as a power change function.

[0061] The discrete points are fitted by the function fitting algorithm in artificial intelligence. The function fitting algorithm that can be used includes linear regression, polynomial regression, or calling function features learned by a deep neural network to fit the function.

[0062] Step S203: Collect the power change functions of all devices in the oil frying production line to obtain the reference function set of the oil frying production line.

[0063] Step S300: Sample the instantaneous power of each device in a certain oil frying production line to obtain a power sampling value sequence, perform cluster analysis on a plurality of power sampling value sequences, and obtain a reference power sequence of a certain oil frying production line.

[0064] The step S300 includes:

[0065] Step S301: Obtain the instantaneous power of all devices in a certain frying production line at a certain time tr, collect the power sample value sequence corresponding to the time tr to obtain the power sample value sequence Qtr, Qtr(p1, p2, p3, …, pn), wherein p1, p2, p3, …, and pn represent the instantaneous power of the 1st, 2nd, 3rd, …, and nth devices in a certain frying production line at the time tr, respectively.

[0066] Step S302: Obtain the power sample value sequences corresponding to a plurality of times of a certain frying production line, convert the power sample value sequences into corresponding value vectors, collect the corresponding vectors of all power sample value sequences into a sample sequence set, and obtain the clustering core of the sample sequence set through a clustering algorithm.

[0067] The high-dimensional vectors are clustered through a distance algorithm in artificial intelligence, and a clustering core is extracted from each sample sequence set. The sampleable clustering algorithms include, for example, a K-means algorithm, a DBSCAN algorithm, or a GMM Gaussian mixture model.

[0068] Step S303: Obtain the corresponding vector of the clustering core, and take the numerical sequence of the corresponding vector as a reference power sequence.

[0069] Step S400: Calculate the power difference evaluation value of the certain frying production line as a target production line.

[0070] The step S400 includes:

[0071] Step S401: Take a certain device in the target production line as a target device, sample the power of the target device i times in a unit period, record a power value each time, and the time length of the unit period is less than or equal to the change period of the target device.

[0072] Step S402: Collect the i power values to form a target power value set, record the jth power value in the target power value set as mj, obtain a reference function f of the target device, draw f in a coordinate system, obtain the coordinates of mj in the coordinate system, and calculate the shortest distance dj of mj to f.

[0073] Step S403: Obtain the shortest distance of each power value in the target power value set to the reference function f, and record the sum of all shortest distances as the power deviation value g of the target device.

[0074] Step S404: Collect the power deviation values of all devices in the target production line, calculate the average of the power deviation values, and obtain the power difference evaluation value W1 of the target production line.

[0075] Step S500: when the power difference is evaluated to exceed the first evaluation threshold, a first type of alarm prompt is performed, the temperature record of the product processed in the target production line in a unit period is collected, and the difference degree of the temperature record and the reference sample is calculated, when the difference degree exceeds the second evaluation threshold, a second type of alarm prompt is performed;

[0076] Step S500 includes:

[0077] Step S501: set the first evaluation threshold H1, when W1 is greater than H1, the first type of alarm is performed on the management personnel of the target production line, and the temperature record of the product after each oiling operation in the unit period is further obtained, and the average value and variance of the product surface temperature after each oiling operation are obtained;

[0078] Step S502: the average value and variance of the product surface temperature after each oiling operation are respectively composed of the target data group corresponding to the oiling operation, all target data groups in the unit period are collected, and the target data sequence is obtained;

[0079] In the embodiment, a certain frying production line includes 5 oiling processes, and the product surface temperature after each oiling process is sampled in a unit period, and each oiling process corresponds to a temperature sequence, wherein the sequences corresponding to the first, second, third, fourth and fifth oiling processes are represented by L1, L2, L3, L4 and L5 respectively;

[0080] Wherein, L1 includes (t11, t12, t13…t1c), t11, t12, t13… and t1c respectively represent the sampling value of the first, second, third, … and cth product surface temperature in the unit period time range;

[0081] The average temperature of L1 is calculated as A1, and the variance of the temperature in L1 is calculated as B2;

[0082] Step S503: obtaining the reference sample of the frying processing equipment in the target production line, and corresponding the oiling process corresponding to each target data group in the target data sequence to the reference data group in the reference sample according to the oiling process;

[0083] Step S504: obtaining the rth target data group (Ar, Br) in the target data sequence, wherein Ar represents the average value of the temperature in the rth target data group, Br represents the variance of the temperature in the rth target data group, obtaining the reference data group (Xr, Yr) corresponding to the rth target data group, calculating the sample difference Ur of the rth target data group, Ur=α×| Ar -Xr|+β×(Br-Yr), wherein α and β are weight coefficients, satisfying the condition α>0, β>0;

[0084] In the embodiment, only the numerical value of the temperature average is taken to bring in the calculation of the sample difference;

[0085] The reference data groups corresponding to the first, second, third, fourth and fifth oil spraying processes are respectively denoted as (60, 8), (100, 8), (140, 6), (160, 4) and (160, 3);

[0086] The corresponding target data groups (58, 8.5), (102, 8), (135, 6), (158, 6) and (165, 3) are collected;

[0087] In the embodiment, α is taken as 0.7 and β is taken as 0.3;

[0088] The calculation results are U1=1.55, U2=1.4, U3=3.5, U4=2.3 and U5=3.5;

[0089] The average value of U1-U5 is calculated to obtain the sample difference 2.45;

[0090] Step S505: The sample differences of all target data groups in the target data sequence are collected, the average value of the sample differences is calculated, and the average value is denoted as the difference degree evaluation value W2;

[0091] Step S506: The second evaluation threshold H2 is set, and when W2 is greater than H2, the second type of alarm is given to the management personnel of the target production line.

[0092] The system comprises a reference sample management module, a reference function management module, a reference power sequence management module, a power difference evaluation module and an information reminding module;

[0093] The reference sample management module is used to collect the characteristic data of the temperature change of the product in the processing process, and the reference sample of the temperature change of the product is obtained by collecting the characteristic data, wherein the reference sample management module comprises a temperature monitoring unit, a reference data group management unit and a sample management unit, the temperature monitoring unit is used to manage the temperature of the product in the processing area, the reference data group management unit is used to manage the reference data group, and the sample management unit is used to collect all reference data groups in the production process under the standard state to obtain the reference sample;

[0094] The reference function management module is configured to obtain the change period of the power of each device, obtain the reference function of the power change of each device, and obtain the reference function set of the frying production line by collecting the reference functions.

[0095] The reference power sequence management module is configured to perform clustering analysis on the power sequence to obtain the reference power sequence of a frying production line.

[0096] The power difference evaluation module is configured to calculate the power difference evaluation value of the target production line.

[0097] The information reminding module is configured to manage the alarm condition and give information prompt when the alarm condition is met.

[0098] It will be apparent to those skilled in the art that the application is not limited to the details of the above-exemplified embodiments and that the present application can be implemented in other particular forms without departing from the spirit or essential characteristics of the present application. The embodiments should therefore be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the above description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. No reference signs in the claims should be considered as limiting the scope of the claims with respect to the figures of the patent document.

Claims

1. An artificial intelligence-based energy-saving management method for a frying production line, characterized by: The method comprises the steps of: Step S100: collecting characteristic data of temperature change of products in the processing process when the frying production line is running in a reference state, and collecting the characteristic data to obtain a reference sample of product temperature change; Step S100 comprises: Step S101: dividing a processing area of a frying processing device in the frying production line into a plurality of unit areas; Step S102: collecting product surface temperature of products in all unit areas after oiling in the production process to obtain an average value and a variance of product temperature after a first oiling operation, and composing the average value and the variance into a reference data group of the first oiling operation; Step S103: collecting all reference data groups in the production process in a certain standard state, arranging the reference data groups in order of oiling procedures to obtain a reference sample; Step S200: collecting historical power records of all devices in the frying production line, obtaining a change period of power of each device, obtaining a reference function of power change of each device, and collecting the reference functions to obtain a reference function set of the frying production line; Step S300: sampling instantaneous power of all devices in a certain frying production line to obtain a power sample value sequence, performing cluster analysis on a plurality of power sample value sequences to obtain a reference power sequence of the certain frying production line; Step S400: calculating a power difference evaluation value for the certain frying production line as a target production line, comprising: Step S401: taking a certain device in the target production line as a target device, sampling power of the target device i times in a unit period, recording a power value each time, and a time length of the unit period is less than or equal to a change period of the target device; Step S402: collecting i power values to compose a target power value set, taking a jth power value in the target power value set as mj, obtaining a reference function f of the target device, drawing f in a coordinate system, obtaining coordinates of mj in the coordinate system, and calculating a shortest distance dj of mj to f; Step S403: obtaining the shortest distances of all power values in the target power value set to the reference function f, and taking a sum of all the shortest distances as a power deviation value g of the target device; Step S404: collecting power deviation values of all devices in the target production line, calculating an average of the power deviation values to obtain a power difference evaluation value W1 of the target production line; Step S500: when the power difference evaluation exceeds a first evaluation threshold, performing a first type of alarm prompt, collecting temperature records of products processed in the target production line in a unit period, calculating a difference degree of the temperature records and the reference sample, and when the difference degree exceeds a second evaluation threshold, performing a second type of alarm prompt.

2. The artificial intelligence-based energy saving management method for a frying line according to claim 1, characterized in that: Step S200 comprises: Step S201: taking a certain device in the frying production line as a reference device, sampling power values of the reference device in a change period, collecting sample values of all samplings to obtain a power sample sequence of the reference device; Step S202: performing function fitting on all sample values in the power sample sequence to obtain a function of power change of the reference device with time, and taking the function as a power change function; Step S203: collect the power change functions of all the devices in the frying production line to obtain a reference function set of the frying production line.

3. The artificial intelligence-based energy saving management method for a frying line according to claim 2, characterized in that: Step S300 includes: Step S301: obtain the instantaneous power of all the devices in the certain frying production line at a time tr, and collect the power sample value sequence Qtr corresponding to the time tr to obtain Qtr (p1, p2, p3, …, pn), where p1, p2, p3, …, and pn represent the instantaneous power of the 1st, 2nd, 3rd, …, and nth devices in the certain frying production line at the time tr, respectively; Step S302: obtain the power sample value sequences corresponding to a plurality of times, respectively, of the certain frying production line, convert the power sample value sequences into corresponding value vectors, collect the corresponding vectors of all the power sample value sequences into a sample sequence set, and obtain the clustering core of the sample sequence set through a clustering algorithm; Step S303: obtain the corresponding vector of the clustering core, and take the value sequence of the corresponding vector as a reference power sequence.

4. The artificial intelligence-based energy-saving management method for a frying production line according to claim 3, characterized in that: Step S500 includes: Step S501: set a first evaluation threshold H1, and when W1 is greater than H1, perform a first type of alarm on the management personnel of the target production line, further obtain the product temperature records after all the oiling operations of the frying processing devices in the target production line in a unit period, and obtain the average value and variance of the product surface temperature after each oiling operation; Step S502: form the average value and variance of the product surface temperature after each oiling operation into a target data group corresponding to the oiling operation, respectively, collect all the target data groups in the unit period to obtain a target data sequence; Step S503: obtain the reference sample of the frying processing devices in the target production line, and correspond the oiling process corresponding to each target data group in the target data sequence to the reference data group in the reference sample according to the oiling process; Step S504: obtain the rth target data group (Ar, Br) in the target data sequence, where Ar represents the average value of the temperature in the rth target data group, and Br represents the variance of the temperature in the rth target data group, obtain the reference data group (Xr, Yr) corresponding to the rth target data group, and calculate the sample difference Ur of the rth target data group, Ur = α × | Ar - Xr | + β × (Br - Yr), where α and β are weight coefficients, and satisfy the conditions α > 0 and β > 0; Step S505: collect the sample differences of all the target data groups in the target data sequence, calculate the average value of the sample differences, and take the average value as a difference degree evaluation value W2; Step S506: set a second evaluation threshold H2, and when W2 is greater than H2, perform a second type of alarm on the management personnel of the target production line.

5. An artificial intelligence-based energy-saving management system for a frying production line, used to execute an artificial intelligence-based energy-saving management method for a frying production line according to any one of claims 1-4, characterized in that: The system includes: The reference sample management module, the reference function management module, the reference power sequence management module, the power difference evaluation module and the information prompting module, wherein the reference sample management module is used for collecting characteristic data of temperature change of products in the processing process, and collecting the characteristic data to obtain reference samples of temperature change of the products, the reference function management module is used for obtaining change periods of power of each device, and obtaining reference functions of power change of each device, and collecting the reference functions to obtain a reference function set of the frying production line, the reference power sequence management module is used for clustering analysis on the power sequence, and obtaining a reference power sequence of the certain frying production line, the power difference evaluation module is used for calculating a power difference evaluation value of the target production line, and the information prompting module is used for managing alarm conditions and prompting information when the alarm conditions are met.

6. The frying production line energy-saving management system based on artificial intelligence according to claim 5, characterized in that: The reference sample management module comprises a temperature monitoring unit, a reference data set management unit and a sample management unit, wherein the temperature monitoring unit is used for managing the temperature of products in the processing area, the reference data set management unit is used for managing reference data sets, and the sample management unit is used for collecting all reference data sets in the production process under a standard state to obtain reference samples; The reference function management module comprises a power sampling sequence management unit, a function fitting unit and a function set management unit, wherein the power sampling sequence management unit is used for collecting sampling values to obtain a power sampling sequence of the reference device, the function fitting unit is used for function fitting on the sampling values in the power sampling sequence to obtain a power change function, and the function set management unit is used for collecting the power change functions to manage the reference function set; The reference power sequence management module comprises a power sampling value sequence management unit, a clustering management unit and a reference sequence management unit, wherein the power sampling value sequence management unit is used for collecting instantaneous power of all devices in the frying production line to manage power sampling value sequences, the clustering management unit is used for clustering on vectors corresponding to the power sampling value sequences to obtain clustering cores, and the reference sequence management unit is used for managing the reference power sequence.

7. The frying production line energy-saving management system based on artificial intelligence according to claim 5, characterized in that: The power difference evaluation module comprises a power sampling unit, a distance management unit, a power deviation value management unit and a power difference evaluation value management unit, wherein the power sampling unit is used for sampling power of the target device in a unit period, the distance management unit is used for managing shortest distances between each power value in a target power value set and the reference function, the power deviation value management unit is used for calculating a power deviation value of the target device, and the power difference evaluation value management unit is used for managing the power difference evaluation value of the target production line.

8. The frying production line energy-saving management system based on artificial intelligence according to claim 5, characterized in that: The information reminding module comprises a first evaluation unit, a target data sequence management unit, a sample difference management unit, a second evaluation unit and an alarm feedback unit, wherein the first evaluation unit is configured to evaluate the value of the power difference evaluation value, the target data sequence management unit is configured to obtain the target data sequence of the frying processing equipment in the target production line, the sample difference management unit is configured to correspond the target data sequence with the reference sample and calculate the sample difference, the second evaluation unit is configured to evaluate the sample difference, and the alarm feedback unit is configured to feed back the alarm information meeting the alarm condition to the relevant management personnel.

Images