Intelligent control system for drying unit

Abstract

The invention relates to the technical field of intelligent control, and discloses an intelligent control system for a drying unit, which comprises a drying unit data acquisition module, a performance analysis module, a target article acquisition module, an article drying attribute acquisition module, a drying unit control strategy acquisition module and an intelligent control module, by arranging the performance analysis module and the article drying attribute acquisition module, the drying units with similar performance can be defined as the same category, and the running performance of the drying unit equipment with the same category is similar, namely, the data such as equipment energy consumption, drying efficiency and drying time are similar when the same article is dried; therefore, when the articles are dried subsequently, the same control strategy can be adopted, and the same control parameters can be obtained, so that the corresponding drying effect is achieved; and the drying attribute of the target object is obtained, so that a foundation is laid for subsequently forming a drying strategy and controlling drying process parameters.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and more specifically to an intelligent control system for a drying unit. Background Art

[0002] A method for intelligent control of heat pump drying unit parameters is disclosed in a public document with publication number CN116577995B, which is applied to an intelligent control system for heat pump drying unit parameters. The method configures N-level drying state cycles according to the basic attributes of the items to be dried; collects data of the items to be dried before drying, clusters the items to be dried based on the data collection results, and generates clustering identifiers; configures N-level drying state cycles as key nodes, and configures fuzzy verification nodes between N-level key nodes; controls the drying of the items to be dried after clustering the items based on the fitting control curve, and performs fuzzy data collection of the items to be dried at the fuzzy verification nodes to obtain the fuzzy data collection results, and performs initial optimization in combination with the fitting result identifier; performs data collection of the items to be dried at the N-level key nodes to obtain the data collection results, and performs correction optimization in combination with the fitting result identifier; performs intelligent drying control of the items to be dried according to the initial optimization results and the correction optimization results. The method solves the technical problem in the prior art that the parameter control accuracy of the heat pump drying unit is low, resulting in poor drying effect of the heat pump drying unit. The technical effect of improving the parameter control accuracy of the heat pump drying unit and thus improving the drying quality of the heat pump drying unit is achieved.

[0003] However, for different items to be dried, although they belong to the same type, the required drying processes are still different, so there are still differences in the parameter requirements for the drying units; and for large-scale drying tasks, multiple drying units are required to work at the same time. If the drying units are not intelligently controlled, they cannot work alternately, which will shorten the service life of the drying units. Long-term uninterrupted operation of the drying units will reduce the drying efficiency and the working performance of the drying units. Therefore, there is a lack of adaptive control over multiple drying units, which makes it impossible to ensure the efficient and stable operation of the drying units, balance the performance status of each drying unit, and optimize the energy consumption resources of the drying units.

[0004] Based on this, the present invention proposes a drying unit and an intelligent control system, which adopts different drying unit operation combinations for different drying strategies, thereby balancing the performance status of each drying unit and extending the service life of each drying unit. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent control system for a drying unit to solve the problems existing in the above-mentioned background technology.

[0006] The present invention provides the following technical solutions: an intelligent control system for a drying unit, comprising a drying unit data acquisition module, a performance analysis module, a target item acquisition module, an item drying attribute acquisition module, a drying unit control strategy acquisition module and an intelligent control module;

[0007] The drying unit data collection module is used to collect the operation control data and drying data of n drying units, set digital tags to mark the n drying units, and transmit them to the performance analysis module;

[0008] The performance analysis module is used to cluster the n drying machine groups and perform performance index analysis on all drying machine groups in each category, and then transmit the performance index to the drying machine group control strategy acquisition module;

[0009] The target object acquisition module is used to acquire target object data and transmit it to the object drying property acquisition module;

[0010] The object drying attribute acquisition module is used to construct an object drying knowledge graph, match the target object to obtain the drying attributes corresponding to the object, and transmit them to the drying unit control strategy acquisition module;

[0011] The drying group control strategy acquisition module uses an optimized genetic algorithm to obtain the optimal operation category of the drying group based on the operation control data and the drying data, combined with the drying properties and performance index of the items, and forms a drying strategy to be transmitted to the intelligent control module;

[0012] The intelligent control module is used to execute the drying strategy and control the operation of the corresponding drying machine group to perform drying operations on the target items.

[0013] Preferably, the operation control data includes a device temperature control value, a device humidity control value and a device operation energy consumption, the drying data is a drying process parameter for each drying of an item, that is, an operation temperature, an operation humidity, a wind speed, a drying time and a drying energy consumption of each drying stage for each drying, the drying effect can be represented by a difference in moisture content before and after drying of the item, and the drying stages include a preheating stage, a constant speed drying stage, a speed reduction drying stage and a cooling stage;

[0014] The target object is the object to be dried, and the target object data includes the name, weight, quantity, and initial moisture content of the target object;

[0015] The drying properties are drying process parameters such as the operating temperature, humidity, corresponding drying time, and drying energy consumption of each drying stage required to achieve the drying state when the number of items is different;

[0016] The drying strategy is the drying process parameters corresponding to the digital labels of each drying unit corresponding to the operating drying unit category and the drying attributes of the items; all drying units of the optimal operating category are used as the drying units for this operation.

[0017] Preferably, the specific manner in which the performance analysis module performs equipment clustering on the n drying machine groups is:

[0018] Based on the operation control data of the drying group, a performance similarity matrix U is constructed, U = comm(a, b), where comm(a, b) represents the performance similarity between the drying group with digital label a and the drying group with digital label b; the formula is expressed as:

[0019]

[0020] Wherein, k1 is a weight parameter, za is a comprehensive feature vector of a drying machine group with a digital label, zb is a comprehensive feature vector of a drying machine group with a digital label b, and the comprehensive feature vector is a vectorized representation of the operation control data; cov(za,zb) is the covariance of za and zb, σ(za) is the standard deviation of za, and σ(zb) is the standard deviation of zb; a=1, 2, 3, ..., n, b=1, 2, 3, ..., n;

[0021] Based on the performance similarity matrix, the drying machine groups are clustered; each drying machine group is regarded as a separate small cluster, and there are n small clusters in total, where n is the total number of drying machine groups; the gravity value between every two small clusters is calculated based on the performance similarity matrix; a merging threshold d is preset to control cluster merging, and the small clusters are iteratively merged. In each iteration, two small clusters q1 and q2 with the largest gravity value are found;

[0022] If the gravity value between q1 and q2 is greater than the merging threshold d, q1 and q2 are merged into a new cluster q, and the gravity value between the new cluster q and all other small clusters is recalculated; otherwise, no iterative merging is performed; the execution is repeated until the gravity value between no two small clusters is greater than the merging threshold d, at which time k new clusters are obtained, that is, k operating performance categories.

[0023] Preferably, the calculation formula of the gravity value is expressed as: Where Y(q1,q2) is the gravitational force between q1 and q2, G is the gravitational constant, c1 is the number of dryer groups in the small cluster q1, c2 is the number of dryer groups in the small cluster q2, Dis(q1,q2) is the distance between the small clusters q1 and q2; that is, the average value of the distance between all pairs of dryer groups in the two small clusters. The distance measurement method between pairs of dryer groups can be calculated using the Euclidean distance formula; ω is a hyperparameter, σ(q1) is the standard deviation of the comprehensive feature vector of the dryer group in the small cluster q1, σ(q2) is the standard deviation of the comprehensive feature vector of the dryer group in the small cluster q2; μ(q1) is the mean of the comprehensive feature vector of the dryer group in the small cluster q1, and μ(q2) is the mean of the comprehensive feature vector of the dryer group in the small cluster q2; comm * (q1,q2) is the mean similarity between the drying machine groups of small clusters q1 and q2.

[0024] Preferably, the performance index analysis of the drying unit comprises the following steps:

[0025] Step S01: Calculate the temperature control performance index: Among them, JD temp,i is the temperature control performance index of the i-th drying unit, N is the total number of times the i-th drying unit is operated, TW ij is the actual temperature of the i-th drying unit during the j-th operation, TW′ ij is the set temperature of the i-th drying unit when it runs for the jth time, j = 1, 2, 3, ..., N; i = 1, 2, 3, ..., n;

[0026] Step S02: Calculate the humidity control performance index: Among them, JD hum,i is the humidity control performance index of the i-th drying unit, N is the number of times the i-th drying unit is run, TS ij is the actual humidity of the i-th drying unit during its j-th operation, TS′ ij is the set humidity of the i-th drying unit during the j-th operation, j = 1, 2, 3, ..., N; i = 1, 2, 3, ..., n;

[0027] Step S03: Calculate the performance index: Among them, δ i is the performance index of the i-th drying unit, θ i is the unit energy consumption of the i-th drying unit, β i are other influencing factors of the i-th drying unit, and λ1, λ2, and λ3 are the corresponding proportional factors.

[0028] Preferably, the method for obtaining the drying attribute corresponding to the item is specifically as follows:

[0029] Collect historical drying item data, and build an item drying knowledge graph based on web crawler technology. The item drying knowledge graph includes entities, entity attribute information, entity relationship information, and entity type.

[0030] Match the target item with the entity in the item drying knowledge graph; the matching methods include string matching and semantic matching; string matching includes exact matching and approximate matching, where exact matching is to compare the string corresponding to the target item with the string corresponding to the entity in the item drying knowledge graph, and if they are exactly the same, then matching is performed; approximate matching is to use a string similarity algorithm to calculate the similarity between the string corresponding to the target item and the string corresponding to the entity in the item drying knowledge graph, and match the entity in the item drying knowledge graph with the highest similarity with the target item; semantic matching is to use word embedding; the word embedding method converts the target item and the entity in the item drying knowledge graph into vector form through a word embedding model, and matches by calculating the similarity between the vectors;

[0031] After the match is successful, the successfully matched entity in the item drying knowledge graph is marked as a matching entity; the matching entities obtained by the two methods are merged, and the entity attribute information corresponding to the matching entity is obtained from the item drying knowledge graph; that is, the drying attributes corresponding to the target item.

[0032] Preferably, the optimal operation category is obtained in the following manner:

[0033] Step S11: Encode k operating performance categories as O, where O is a chromosome, obtain the chromosome, and construct an initial population A = {O1, O2, O3, ..., O k};

[0034] Step S12: determining a fitness function;

[0035] Step S13: performing natural selection on chromosomes in the population;

[0036] Step S14: performing crossover recombination on chromosomes in the population;

[0037] Step S15: mutating the chromosomes in the population;

[0038] Step S16: Obtain a new population, preset the population generation to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps S13 to S15 until the generation corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the operating performance category corresponding to the chromosome with the maximum fitness in the new population is obtained, which is the optimal operating category.

[0039] Preferably, the fitness function is expressed as: r =-ρ r , where f r is the fitness corresponding to the rth chromosome, ρ r The predicted performance attenuation value of the drying machine group of the operating performance category corresponding to the rth chromosome; r = 1, 2, 3, ..., k;

[0040] The method for obtaining the predicted value of the performance attenuation of the drying machine group is as follows: the performance index of each drying machine group in the operating performance category corresponding to the chromosome and the corresponding operating control data and drying data are used as research data, the research data are input into the trained performance prediction model, and the performance index corresponding to each drying machine group is predicted; the performance index of the last operation is subtracted from the predicted performance index to obtain the performance attenuation value of each drying machine group, and the performance attenuation values ​​of all drying machines in the operating performance category are averaged as the predicted performance attenuation value of the operating performance category;

[0041] The training method of the performance prediction model is specifically as follows:

[0042] The operation control data and drying data corresponding to a drying machine group are taken as an analysis data set, and the operation control data and drying data of each operation are taken as an element in an analysis data set. E analysis data sets are collected in advance, and the performance index corresponding to each operation is obtained for the elements in the e analysis data sets, where e is an integer greater than 1, and the elements in each analysis data set and the corresponding performance index are converted into a corresponding set of feature vectors; e = 1, 2, 3, ..., n;

[0043] Each set of feature vectors is used as the input of the performance prediction model. The performance prediction model uses a predicted performance index corresponding to each element in the analysis data set as the output, and uses the actual performance index value corresponding to each element in the analysis data set as the prediction target. The actual performance index value can be obtained through the performance index analysis process; the training target is to minimize the sum of the prediction errors of the elements in all the analysis data sets; the formula of the prediction error is expressed as: ε p =γ p -μ p , where ε p is the prediction error, p is the group number of the eigenvector corresponding to the element in the analysis data set, γ p is the prediction performance index corresponding to the elements in the p-th group of analysis data set, μ p The actual performance index corresponding to the elements in the p-th group of analysis data set is used to train the performance prediction model until the sum of the prediction errors reaches convergence and the training is stopped;

[0044] The performance prediction model is specifically a deep network model.

[0045] Technical effects and advantages of the present invention:

[0046] (1) The present invention is provided with a performance analysis module and an item drying property acquisition module, which is conducive to defining drying units with similar performance as the same category. The operating performance of drying units of the same category is similar, that is, the equipment energy consumption, drying efficiency, drying time and other data when drying the same item are similar. Although drying units of different categories adopt the same control strategy, due to different equipment performance, the equipment energy consumption and drying effects are also different, which will cause uneven drying of the items. Therefore, when the items are subsequently dried, the same control strategy can be adopted and the same control parameters can be obtained to achieve the corresponding drying effect. Obtaining the drying properties of the target item lays the foundation for the subsequent formation of a drying strategy and the control of the drying process parameters.

[0047] (2) The present invention is provided with a drying unit control strategy acquisition module, which is conducive to integrating the trained neural network model into the fitness calculation of the optimized genetic algorithm by adopting the optimized genetic algorithm, and can give full play to the nonlinear and multimodal operation control data and drying data modeling capabilities of the neural network, and effectively capture complex data relationships; based on the parallel computing processing mechanism, the computing efficiency is improved, and the operation performance of the drying unit is evaluated through deep learning technology, thereby improving the pertinence and effectiveness of extending the service life of the drying unit; fully utilizing the operation control data and drying data, and capturing complex data relationships through deep learning technology, it can comprehensively and accurately understand the changes in the equipment operation performance of different drying units, and provide customized operation plans for various types of drying units, thereby effectively improving the operation performance of the overall drying unit, extending the service life of the drying unit, and ensuring that the overall drying unit is maintained at the best performance to achieve performance balance. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The figure is a flow chart of the intelligent control system for a drying unit of the present invention. DETAILED DESCRIPTION

[0049] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are merely illustrative. The intelligent control system for a drying unit involved in the present invention is not limited to the various structures recorded in the following embodiments. All other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0050] like Figure 1As shown, the present invention provides an intelligent control system for a drying group, including a drying group data acquisition module, a performance analysis module, a target item acquisition module, an item drying attribute acquisition module, a drying group control strategy acquisition module and an intelligent control module;

[0051] The drying unit data acquisition module is used to collect the operation control data and drying data of n drying units, set digital tags for the n drying units to mark them, and transmit them to the performance analysis module; the operation control data includes but is not limited to the equipment temperature control value, the equipment humidity control value and the equipment operation energy consumption, etc., the drying data is the drying process parameters when drying items each time, that is, the operation temperature, operation humidity, wind speed, drying time and drying energy consumption of each drying stage each time, the drying effect can be represented by the difference in moisture content before and after drying of the items, if the difference in moisture content before and after drying is larger, the drying effect is better, if the difference in moisture content before and after drying is smaller, the drying effect is worse; the drying stages include a preheating stage, a constant speed drying stage, a speed reduction drying stage and a cooling stage;

[0052] The performance analysis module is used to cluster the n drying groups and perform performance index analysis on all drying groups in each category, and then transmit the results to the drying group control strategy acquisition module; the purpose is to make the operating performance of the drying group equipment of the same category similar, that is, the equipment energy consumption, drying efficiency, drying time and other data when the same items are dried are similar, so that when the items are subsequently dried, the same control strategy can be adopted and the same control parameters can be obtained, so as to achieve the corresponding drying effect; although the drying groups of different categories adopt the same control strategy, due to different equipment performance, the equipment energy consumption and drying effect are also different, which will lead to uneven drying of the items;

[0053] The target item acquisition module is used to acquire target item data and transmit it to the item drying property acquisition module; the target item is the item to be dried, and the target item data includes but is not limited to basic information such as the name, weight, quantity, and initial moisture content of the target item;

[0054] The item drying attribute acquisition module is used to construct an item drying knowledge graph, match the target item to obtain the drying attributes corresponding to the item, and transmit them to the drying unit control strategy acquisition module; the drying attributes are the drying process parameters such as the operating temperature, humidity and corresponding drying time of each drying stage required to achieve the drying state when the number of items is different, as well as the drying energy consumption;

[0055] The drying unit control strategy acquisition module uses an optimized genetic algorithm to obtain the optimal operation category of the drying unit based on the operation control data and drying data, combined with the drying properties and performance index of the items, and forms a drying strategy to be transmitted to the intelligent control module; all drying units of the optimal operation category are used as the drying units in this operation; the drying strategy is the drying process parameters corresponding to the digital labels of each drying unit corresponding to the operating drying unit category and the drying properties of the items; its purpose is to integrate the trained neural network model into the fitness calculation of the optimized genetic algorithm by using the optimized genetic algorithm, so as to give full play to the nonlinear and multimodal operation of the neural network The control data and drying data modeling capabilities can effectively capture complex data relationships; based on the parallel computing processing mechanism, the computing efficiency is improved, and the operating performance of the drying unit is evaluated through deep learning technology, thereby improving the pertinence and effectiveness of extending the service life of the drying unit; making full use of the operating control data and drying data, and capturing complex data relationships through deep learning technology, it can fully and accurately understand the changes in the equipment operating performance of different drying units, and provide customized operating solutions for various types of drying units, thereby effectively improving the operating performance of the overall drying unit, extending the service life of the drying unit, and ensuring that the overall drying unit is maintained at the best performance and achieving performance balance;

[0056] The intelligent control module is used to execute the drying strategy, control the operation of the corresponding drying unit to dry the target object, and control the temperature, humidity, wind speed and other parameters of each drying unit.

[0057] In this embodiment, it should be specifically explained that the specific manner in which the performance analysis module performs equipment clustering on the n drying machine groups is as follows:

[0058] Based on the operation control data of the drying group, a performance similarity matrix U is constructed, U = comm(a, b), where comm(a, b) represents the performance similarity between the drying group with digital label a and the drying group with digital label b; the formula is expressed as:

[0059]

[0060] Wherein, k1 is a weight parameter, za is a comprehensive feature vector of a drying machine group with a digital label, zb is a comprehensive feature vector of a drying machine group with a digital label b, and the comprehensive feature vector is a vectorized representation of the operation control data; cov(za,zb) is the covariance of za and zb, σ(za) is the standard deviation of za, and σ(zb) is the standard deviation of zb; a=1, 2, 3, ..., n, b=1, 2, 3, ..., n;

[0061] Based on the performance similarity matrix, the drying machine groups are clustered; specifically, each drying machine group is regarded as a separate small cluster, and there are n small clusters in total, where n is the total number of drying machine groups; the gravity value between each two small clusters is calculated based on the performance similarity matrix; a merging threshold d is preset to control cluster merging, and the small clusters are iteratively merged. In each iteration, two small clusters q1 and q2 with the largest gravity value are found;

[0062] If the gravity value between q1 and q2 is greater than the merging threshold d, q1 and q2 are merged into a new cluster q, and the gravity value between the new cluster q and all other small clusters is recalculated; otherwise, no iterative merging is performed; the execution is repeated until the gravity value between no two small clusters is greater than the merging threshold d, at which time k new clusters are obtained, that is, k operating performance categories; the performance of drying groups in the same category is similar, so the same drying group control strategy can be adopted and the same control parameters can be obtained, thereby ensuring that the items achieve the corresponding drying effect, laying the foundation for the subsequent selection of drying groups in the same operating performance category for operation control.

[0063] In this embodiment, it should be specifically explained that the calculation formula of the gravity value is expressed as: Where Y(q1,q2) is the gravitational force between q1 and q2, G is the gravitational constant, c1 is the number of dryer groups in the small cluster q1, c2 is the number of dryer groups in the small cluster q2, Dis(q1,q2) is the distance between the small clusters q1 and q2; that is, the average value of the distance between all pairs of dryer groups in the two small clusters. The distance measurement method between pairs of dryer groups can be calculated using the Euclidean distance formula; ω is a hyperparameter used to control the overall size; σ(q1) is the standard deviation of the comprehensive feature vector of the dryer group in the small cluster q1, and σ(q2) is the standard deviation of the comprehensive feature vector of the dryer group in the small cluster q2, reflecting the degree of dispersion within the small cluster; μ(q1) is the mean of the comprehensive feature vector of the dryer group in the small cluster q1, and μ(q2) is the mean of the comprehensive feature vector of the dryer group in the small cluster q2, reflecting the overall level of the small cluster; comm * (q1,q2) is the mean similarity between the drying machine groups of small clusters q1 and q2;

[0064] By adjusting the size of the merge threshold d, the strictness of the merge can be controlled, thereby affecting the final number of clusters. The size of the merge threshold d needs to be determined based on the actual equipment operation and energy consumption, so that a reasonable merge threshold is finally obtained to obtain the ideal number of clusters k.

[0065] In this embodiment, it should be specifically explained that the performance index analysis of the drying unit includes the following steps:

[0066] Step S01: Calculate the temperature control performance index: Among them, JD temp,i is the temperature control performance index of the i-th drying unit, N is the total number of times the i-th drying unit is operated, TW ij is the actual temperature of the i-th drying unit during the j-th operation, TW′ ij is the set temperature of the i-th drying unit when it runs for the jth time, j = 1, 2, 3, ..., N; i = 1, 2, 3, ..., n;

[0067] Step S02: Calculate the humidity control performance index: Among them, JD hum,i is the humidity control performance index of the i-th drying unit, N is the number of times the i-th drying unit is run, TS ij is the actual humidity of the i-th drying unit during its j-th operation, TS′ ij is the set humidity of the i-th drying unit during the j-th operation, j = 1, 2, 3, ..., N; i = 1, 2, 3, ..., n;

[0068] Step S03: Calculate the performance index: Among them, δ i is the performance index of the i-th drying unit, θ i is the unit energy consumption of the i-th drying unit, β i is the other influencing factor of the i-th drying unit, λ1, λ2, and λ3 are the corresponding proportional factors, which are all greater than 0 and less than 1. At the same time, λ1 and λ2 are greater than λ3. For the drying unit, humidity and temperature are the most important control parameters, so the corresponding proportional factors are greater than the proportional factors corresponding to other control parameters. The other influencing factors are the performance index of other control parameters of the drying unit, and the calculation method is the same as the calculation method of the temperature control performance index and the humidity control performance index. For example, if the other influencing factor is the wind speed performance index, the calculation formula of the wind speed performance index is: Among them, JD wind,i is the wind speed control performance index of the i-th drying unit, N is the number of times the i-th drying unit is run, TF ij is the actual wind speed of the i-th drying unit during its j-th operation, TF′ ij is the set wind speed of the i-th drying unit when it is running for the jth time, j=1, 2, 3, ..., N; i=1, 2, 3, ..., n; other influencing factors can be specifically selected by those skilled in the art according to the specific model of the drying unit and the actual drying situation;

[0069] By analyzing the performance index of the drying group, the performance index changes of n drying groups can be obtained, laying the foundation for the subsequent optimization of the genetic algorithm.

[0070] In this embodiment, it should be specifically explained that the method for obtaining the drying attribute corresponding to the item is as follows:

[0071] Collect historical drying item data, and build an item drying knowledge graph based on web crawler technology. The item drying knowledge graph includes entities, entity attribute information, entity relationship information, and entity type, etc. Entities are drying items, which are items that have been dried by the drying unit in the past and items that have not been dried by the drying unit but can be dried by the drying unit; historical drying item data are item data dried by the drying unit, including item name, drying weight, quantity, drying effect, drying process, etc.; entity attribute information is entity drying attribute, and entity type defines the item category to which the entity belongs, including clothing, food, medicine, etc., and defines the general drying attributes and relationships between entities; entity relationship information represents the semantic connection between different entities, and describes various associations and interactions between entities in the knowledge graph;

[0072] Match the target item with the entity in the item drying knowledge graph; the matching methods include string matching and semantic matching; string matching includes exact matching and approximate matching, where exact matching is to compare the string corresponding to the target item with the string corresponding to the entity in the item drying knowledge graph, and if they are exactly the same, then matching is performed; approximate matching is to use a string similarity algorithm to calculate the similarity between the string corresponding to the target item and the string corresponding to the entity in the item drying knowledge graph, and match the entity in the item drying knowledge graph with the highest similarity with the target item; semantic matching is to use word embedding; the word embedding method converts the target item and the entity in the item drying knowledge graph into vector form through a word embedding model, and matches by calculating the similarity between the vectors;

[0073] After the match is successful, the successfully matched entity in the item drying knowledge graph is marked as a matching entity; the matching entities obtained by the two methods are merged, and the entity attribute information corresponding to the matching entity is obtained from the item drying knowledge graph; that is, the drying attributes corresponding to the target item.

[0074] In this embodiment, it should be specifically explained that the method for obtaining the optimal operation category is specifically as follows:

[0075] Step S11: Encode k operating performance categories as O, where O is a chromosome, obtain the chromosome, and construct an initial population A = {O1, O2, O3, ..., O k};

[0076] Step S12: determining a fitness function;

[0077] Step S13: performing natural selection on chromosomes in the population;

[0078] Step S14: performing crossover recombination on chromosomes in the population;

[0079] Step S15: mutating the chromosomes in the population;

[0080] Step S16: obtain a new population, the preset population generation number is L, the fitness threshold is Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps S13 to S15 until the generation number corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the operation performance category corresponding to the chromosome with the maximum fitness in the new population is obtained, which is the optimal operation category; illustratively, if the preset population generation number is 1, the chromosomes in the initial population are subjected to natural selection, crossover recombination, and mutation to obtain a new population, and the generation number corresponding to the new population is 1 at this time, so the loop ends;

[0081] The fitness threshold Q is preset by those skilled in the art according to the algorithm accuracy, and the population algebra L is obtained by those skilled in the art using a genetic algorithm multiple times under multiple sets of different operation control data, drying data and drying attributes to obtain the corresponding optimal operation category. In each process of using the genetic algorithm, when the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends and the algebra corresponding to the new population is obtained; the largest algebra among the multiple algebras is taken as the population algebra L.

[0082] In this embodiment, it should be specifically explained that the expression of the fitness function is: r =-ρ r , where f r is the fitness corresponding to the rth chromosome, ρ r The predicted performance attenuation value of the drying machine group of the operating performance category corresponding to the rth chromosome; r = 1, 2, 3, ..., k;

[0083] The method for obtaining the predicted value of the performance attenuation of the drying machine group is as follows: the performance index of each drying machine group in the operating performance category corresponding to the chromosome and the corresponding operating control data and drying data are used as research data, the research data are input into the trained performance prediction model, and the performance index corresponding to each drying machine group is predicted; the performance index of the last operation is subtracted from the predicted performance index to obtain the performance attenuation value of each drying machine group, and the performance attenuation values ​​of all drying machines in the operating performance category are averaged as the predicted performance attenuation value of the operating performance category;

[0084] The training method of the performance prediction model is specifically as follows:

[0085] The operation control data and drying data corresponding to a drying machine group are taken as an analysis data set, and the operation control data and drying data of each operation are taken as an element in an analysis data set. E analysis data sets are collected in advance, and the performance index corresponding to each operation is obtained for the elements in the e analysis data sets, where e is an integer greater than 1, and the elements in each analysis data set and the corresponding performance index are converted into a corresponding set of feature vectors; e = 1, 2, 3, ..., n;

[0086] Each set of feature vectors is used as the input of the performance prediction model. The performance prediction model uses a predicted performance index corresponding to each element in the analysis data set as the output, and uses the actual performance index value corresponding to each element in the analysis data set as the prediction target. The actual performance index value can be obtained through the performance index analysis process; the training target is to minimize the sum of the prediction errors of the elements in all the analysis data sets; the formula of the prediction error is expressed as: ε p =γ p -μ p , where ε p is the prediction error, p is the group number of the eigenvector corresponding to the element in the analysis data set, γ p is the prediction performance index corresponding to the elements in the p-th group of analysis data set, μ p The actual performance index corresponding to the elements in the p-th group of analysis data set is used to train the performance prediction model until the sum of the prediction errors reaches convergence and the training is stopped;

[0087] The performance prediction model is specifically a deep network model.

[0088] In this embodiment, it should be specifically explained that the natural selection is carried out by combining the elite method and the rotation method; the elite method produces F1 offspring chromosomes, and for a population with a capacity of k, the fitness corresponding to the k chromosomes is arranged from large to small, and each of the F1 chromosomes at the front produces a offspring chromosome; the rotation method produces F2 offspring chromosomes, that is, k chromosomes produce F2 offspring chromosomes according to the corresponding rotation probability; F1+F2=k, so as to keep the offspring population capacity k unchanged and the population generation increases;

[0089] The expression of the rotation probability is: Among them, r is the rotation probability corresponding to the rth chromosome;

[0090] In the step S14, E chromosomes are randomly selected from the population for crossover recombination to obtain E new chromosomes; the crossover recombination adopts the PMX method, which is a prior art method and is not described in detail in this embodiment; after the chromosome crossover recombination, the fitness of the E new chromosomes is calculated, and the fitness of the E new chromosomes and the fitness of the E chromosomes are sorted from large to small to generate a sorting table, and the E new chromosomes in the sorting table are replaced with the E chromosomes for crossover recombination in the population in positive order; in this embodiment, E=0.7k is preferred, and if the calculated E is not an integer, E is rounded up to ensure that the calculated E is an integer;

[0091] In step S15, the mutation probability is preset to V, and the k chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes. In this embodiment, V=0.02 is preferred. The mutation probability is preset by technicians in this field according to the algorithm efficiency and algorithm accuracy.

[0092] In this embodiment, it should be specifically explained that the drying strategy is obtained based on the optimal operating category and the drying attributes of the items. The optimal operating category is the drying units that perform the drying operation this time. The drying attributes of the items can obtain the drying process parameters, namely the drying time, drying temperature and drying humidity of each stage. After the intelligent control module executes the drying strategy, it operates the drying units in the optimal category and adjusts the drying process parameters.

[0093] Finally: 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 principles of the present invention should be included in the protection scope of the present invention.

[0094] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent control system for a drying unit, characterized in that: It includes a drying unit data acquisition module, a performance analysis module, a target item acquisition module, an item drying attribute acquisition module, a drying unit control strategy acquisition module and an intelligent control module; The drying unit data collection module is used to collect the operation control data and drying data of n drying units, set digital tags to mark the n drying units, and transmit them to the performance analysis module; The performance analysis module is used to cluster the n drying machine groups and perform performance index analysis on all drying machine groups in each category, and then transmit the performance index to the drying machine group control strategy acquisition module; The target object acquisition module is used to acquire target object data and transmit it to the object drying property acquisition module; The object drying attribute acquisition module is used to construct an object drying knowledge graph, match the target object to obtain the drying attributes corresponding to the object, and transmit them to the drying unit control strategy acquisition module; The drying group control strategy acquisition module uses an optimized genetic algorithm to obtain the optimal operation category of the drying group based on the operation control data and the drying data, combined with the drying properties and performance index of the items, and forms a drying strategy to be transmitted to the intelligent control module; The intelligent control module is used to execute the drying strategy and control the operation of the corresponding drying machine group to perform drying operations on the target items.

2. The intelligent control system for a drying unit according to claim 1, characterized in that: The operation control data includes the equipment temperature control value, the equipment humidity control value and the equipment operation energy consumption. The drying data is the drying process parameters when drying items each time, that is, the operation temperature, operation humidity, wind speed, drying time and drying energy consumption of each drying stage each time. The drying effect can be represented by the difference in moisture content before and after drying of the items. The drying stages include a preheating stage, a constant speed drying stage, a reduced speed drying stage and a cooling stage. The target object is the object to be dried, and the target object data includes the name, weight, quantity, and initial moisture content of the target object; The drying properties are drying process parameters such as the operating temperature, humidity, corresponding drying time, and drying energy consumption of each drying stage required to achieve the drying state when the number of items is different; The drying strategy is the drying process parameters corresponding to the digital labels of each drying unit corresponding to the operating drying unit category and the drying attributes of the items; all drying units of the optimal operating category are used as the drying units for this operation.

3. The intelligent control system for a drying unit according to claim 1, characterized in that: The specific method of the performance analysis module to cluster the n drying machine groups is as follows: Based on the operation control data of the drying group, a performance similarity matrix U is constructed, U = comm(a, b), where comm(a, b) represents the performance similarity between the drying group with digital label a and the drying group with digital label b; the formula is expressed as: Wherein, k1 is a weight parameter, za is a comprehensive feature vector of a drying machine group with a digital label, zb is a comprehensive feature vector of a drying machine group with a digital label b, and the comprehensive feature vector is a vectorized representation of the operation control data; cov(za,zb) is the covariance of za and zb, σ(za) is the standard deviation of za, and σ(zb) is the standard deviation of zb; a=1, 2, 3, ..., n, b=1, 2, 3, ..., n; Based on the performance similarity matrix, the drying machine groups are clustered; each drying machine group is regarded as a separate small cluster, and there are n small clusters in total, where n is the total number of drying machine groups; the gravity value between every two small clusters is calculated based on the performance similarity matrix; a merging threshold d is preset to control cluster merging, and the small clusters are iteratively merged. In each iteration, two small clusters q1 and q2 with the largest gravity value are found; If the gravity value between q1 and q2 is greater than the merging threshold d, q1 and q2 are merged into a new cluster q, and the gravity value between the new cluster q and all other small clusters is recalculated; otherwise, no iterative merging is performed; the execution is repeated until the gravity value between no two small clusters is greater than the merging threshold d, at which time k new clusters are obtained, that is, k operating performance categories.

4. The intelligent control system for a drying unit according to claim 3, characterized in that: The calculation formula of the gravity value is expressed as: Where Y(q1,q2) is the gravitational force between q1 and q2, G is the gravitational constant, c1 is the number of dryer groups in the small cluster q1, c2 is the number of dryer groups in the small cluster q2, Dis(q1,q2) is the distance between the small clusters q1 and q2; that is, the average value of the distance between all pairs of dryer groups in the two small clusters. The distance measurement method between pairs of dryer groups can be calculated using the Euclidean distance formula; ω is a hyperparameter, σ(q1) is the standard deviation of the comprehensive feature vector of the dryer group in the small cluster q1, σ(q2) is the standard deviation of the comprehensive feature vector of the dryer group in the small cluster q2; μ(q1) is the mean of the comprehensive feature vector of the dryer group in the small cluster q1, and μ(q2) is the mean of the comprehensive feature vector of the dryer group in the small cluster q2; comm * (q1,q2) is the mean similarity between the drying machine groups of small clusters q1 and q2.

5. The intelligent control system for a drying unit according to claim 4, characterized in that: The performance index analysis of the drying unit comprises the following steps: Step S01: Calculate the temperature control performance index: Among them, JD temp,i is the temperature control performance index of the i-th drying unit, N is the total number of times the i-th drying unit is operated, TW ij is the actual temperature of the i-th drying unit during the j-th operation, TW′ ij is the set temperature of the i-th drying unit when it runs for the jth time, j = 1, 2, 3, ..., N; i = 1, 2, 3, ..., n; Step S02: Calculate the humidity control performance index: Among them, JD hum,i is the humidity control performance index of the i-th drying unit, N is the number of times the i-th drying unit is run, TS ij is the actual humidity of the i-th drying unit during its j-th operation, TS′ ij is the set humidity of the i-th drying unit during the j-th operation, j = 1, 2, 3, ..., N; i = 1, 2, 3, ..., n; Step S03: Calculate the performance index: Among them, δ i is the performance index of the i-th drying unit, θ i is the unit energy consumption of the i-th drying unit, β i are other influencing factors of the i-th drying unit, and λ1, λ2, and λ3 are the corresponding proportional factors.

6. The intelligent control system for a drying unit according to claim 5, characterized in that: The specific method for obtaining the drying attribute corresponding to the item is: Collect historical drying item data, and build an item drying knowledge graph based on web crawler technology. The item drying knowledge graph includes entities, entity attribute information, entity relationship information, and entity type. Match the target item with the entity in the item drying knowledge graph; the matching methods include string matching and semantic matching; string matching includes exact matching and approximate matching, the exact matching is to compare the string corresponding to the target item with the string corresponding to the entity in the item drying knowledge graph, if they are exactly the same, then match; approximate matching is to use the string similarity algorithm to calculate the similarity between the string corresponding to the target item and the string corresponding to the entity in the item drying knowledge graph, and match the entity in the item drying knowledge graph with the highest similarity with the target item; semantic matching is to use word embedding; the word embedding method converts the target item and the entity in the item drying knowledge graph into vector form through the word embedding model, and matches by calculating the similarity between the vectors; After the match is successful, the successfully matched entity in the item drying knowledge graph is marked as a matching entity; the matching entities obtained by the two methods are merged, and the entity attribute information corresponding to the matching entity is obtained from the item drying knowledge graph; that is, the drying attributes corresponding to the target item.

7. The intelligent control system for a drying unit according to claim 6, characterized in that: The method for obtaining the optimal operation category is specifically as follows: Step S11: Encode k operating performance categories as O, where O is a chromosome, obtain the chromosome, and construct an initial population A = {O1, O2, O3, ..., O k }; Step S12: determining the fitness function; Step S13: performing natural selection on chromosomes in the population; Step S14: performing crossover recombination on chromosomes in the population; Step S15: mutating the chromosomes in the population; Step S16: Obtain a new population, preset the population generation to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps S13 to S15 until the generation corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the operating performance category corresponding to the chromosome with the maximum fitness in the new population is obtained, which is the optimal operating category.

8. The intelligent control system for a drying unit according to claim 7, characterized in that: The fitness function is expressed as: r =-ρ r , where f r is the fitness corresponding to the rth chromosome, ρ r The predicted performance attenuation value of the drying machine group of the operating performance category corresponding to the rth chromosome; r = 1, 2, 3, ..., k; The method for obtaining the predicted value of the performance attenuation of the drying machine group is as follows: the performance index of each drying machine group in the operating performance category corresponding to the chromosome and the corresponding operating control data and drying data are used as research data, the research data are input into the trained performance prediction model, and the performance index corresponding to each drying machine group is predicted; the performance index of the last operation is subtracted from the predicted performance index to obtain the performance attenuation value of each drying machine group, and the performance attenuation values ​​of all drying machines in the operating performance category are averaged as the predicted performance attenuation value of the operating performance category; The training method of the performance prediction model is specifically as follows: The operation control data and drying data corresponding to a drying machine group are taken as an analysis data set, and the operation control data and drying data of each operation are taken as an element in an analysis data set. E analysis data sets are collected in advance, and the performance index corresponding to each operation is obtained for the elements in the e analysis data sets, where e is an integer greater than 1, and the elements in each analysis data set and the corresponding performance index are converted into a corresponding set of feature vectors; e = 1, 2, 3, ..., n; Each set of feature vectors is used as the input of the performance prediction model. The performance prediction model uses a predicted performance index corresponding to each element in the analysis data set as the output, and uses the actual performance index value corresponding to each element in the analysis data set as the prediction target. The actual performance index value can be obtained through the performance index analysis process; the training target is to minimize the sum of the prediction errors of the elements in all the analysis data sets; the formula of the prediction error is expressed as: ε p =γ p -μ p , where ε p is the prediction error, p is the group number of the eigenvector corresponding to the element in the analysis data set, γ p is the prediction performance index corresponding to the elements in the p-th group of analysis data set, μ p The actual performance index corresponding to the elements in the p-th group of analysis data set is used to train the performance prediction model until the sum of the prediction errors reaches convergence and the training is stopped; The performance prediction model is specifically a deep network model.

Images