Intelligent industrial control system and method for electrostatic spraying of aluminum veneer

Abstract

The invention discloses an intelligent industrial control system and method for electrostatic spraying of an aluminum veneer, and relates to the technical field of spraying control, and the control method comprises the following steps: generating a corresponding spraying record for each spraying process of the aluminum veneer; performing multi-dimensional quality evaluation to obtain a quality evaluation value of any spraying record; performing category division on the dimension data of each dimension to respectively obtain feature sets under different category dimensions; analyzing the difference condition of quality evaluation of each feature in the same category in different spraying records to obtain the correlation degree between the features in the same category; analyzing deviation conditions of different types of features in different spraying records to obtain influence degrees among different types of features; when the aluminum veneer is sprayed, data acquisition is synchronously carried out, a real-time evaluation value at the current moment is obtained, and abnormity identification is carried out; and for an existing abnormal condition, making a corresponding adjustment strategy to adjust each abnormal feature.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial control. Background Art

[0002] Electrostatic spraying utilizes the principle of a high-voltage electrostatic field to charge paint particles, which then adsorb evenly onto the surface of the aluminum veneer. This method effectively reduces paint waste and improves coating uniformity and adhesion. As a core surface treatment process in the fields of construction, automotive, and electronics, the quality of electrostatic spraying directly impacts the product's weather resistance, aesthetics, and functionality. Traditional control systems achieve coating through fixed spraying parameters. However, in the actual production process, the spraying quality is easily affected by external environmental factors such as temperature, humidity, and dust, and internal factors such as paint quality and spray gun performance, which can lead to coating defects. However, workers often control the influencing factors in isolation without considering the interaction between different influencing factors, which ultimately leads to the repair process not meeting expectations or causing a large amount of waste of resources. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent industrial control system and method for electrostatic spraying of aluminum veneer to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solution: an intelligent industrial control method for electrostatic spraying of aluminum veneer, the control method comprising the following steps: Step S100: performing multi-dimensional data collection on each spraying process of the aluminum veneer to generate corresponding spraying records; performing multi-dimensional quality assessment on any spraying record to obtain a quality assessment value of any spraying record; Step S200: classifying the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; analyzing the dimensional data of different categories of spraying records in each dimension to obtain feature sets in different categories and dimensions; Step S300: Analyze the differences in quality assessment of each feature in the same category in different spraying records to obtain the correlation degree between features in the same category; Step S400: Analyze the deviation of different types of features in different spraying records to obtain the degree of influence between different types of features; Step S500: Whenever the aluminum veneer is sprayed, data is collected synchronously to obtain the real-time evaluation value at the current moment and perform abnormality identification; for any abnormal situation, a corresponding adjustment strategy is formulated to adjust each abnormal feature.

[0005] Furthermore, step S100 includes the following steps: Step S101: Several monitoring devices are installed to continuously monitor and collect data during the spraying process of the aluminum veneer, and the data type of the monitoring data is obtained and used as the dimension of the monitoring device; all monitoring data collected by any monitoring device are obtained to obtain a dimensional data set of the corresponding dimension, and the dimensional data set of each monitoring device is summarized to generate a spraying record; Step S102: randomly select a spraying record, obtain the dimension data of each dimension in the spraying record, preset corresponding evaluation rules for each dimension, evaluate the dimension data of each dimension to obtain the evaluation value of the dimension; preset an influence weight for each dimension, set the evaluation value of the i-th dimension in the spraying record to p i and the influence weight is μ i , according to the formula: ; Wherein, i is a positive integer and i∈(1,a), a is the number of dimensions in the spraying record; the quality evaluation value Z of the spraying record is calculated; the preset evaluation rules, such as the temperature dimension in the environmental factors, evaluate the degree of deviation and stability of the temperature fluctuation amplitude of the spraying environment, the stability of the temperature will directly affect the spraying quality, or whether the spray flow of the spray gun meets the expectations, the evaluation of the deviation amplitude can also reflect the spraying quality; different settings for different dimensions are different because the impact of different dimensions will be different. If a simple average value calculation is performed, it will cause deviations in the quality evaluation results. The setting of the impact weight can be obtained through the actual deviation of the historical spraying records.

[0006] Furthermore, step S200 includes the following steps: Step S201: Pre-build an equipment dimension database, which contains several dimension categories. Select any one dimension from the several dimensions in the spraying record, and compare the selected dimension with each dimension category to obtain the similarity between the selected dimension and each dimension category; select the dimension category with the greatest similarity and set the maximum similarity as S. max , preset a similarity threshold S th , if S max ≥S th , then the selected dimension is set as the first dimension type set. If S max <S th , the selected dimension is set as the second dimension type set; the first dimension type set represents properties related to the spray gun or aluminum veneer, such as spray flow rate, spray distance, and aluminum veneer flatness, while the second dimension type set represents properties related to external environmental factors, such as temperature and humidity. Step S202: Preset a quality assessment threshold Z th, if there is a spray record that satisfies Z<Z th , then the corresponding spray record is set as an abnormal spray record. If Z≥Z th , it is set as a normal spraying record; arbitrarily select a normal spraying record, and arbitrarily select the dimension data set of the i-th dimension from the selected normal spraying record. If the i-th dimension belongs to the first dimension type set, the value range of the dimension data set is obtained; obtain the value range of the i-th dimension in each normal spraying record and perform a union operation to obtain the normal value range of the i-th dimension [D i (min),D i (max)]; Step S203: Set two sets of feature sets for the i-th dimension, and set the dimension data of the i-th dimension to D i , when D is satisfied i <D i (min), then one of the feature sets is selected as the dimension data D i The feature set of i >D i (max), another set of feature sets is selected as the dimension data D i The value of any dimension may have abnormalities at both ends. For example, the humidity dimension may be too dry or too wet, and the effects will be different. Therefore, different features may be generated even in the same dimension. Step S204: randomly select an abnormal spraying record and obtain the value range of the i-th dimension in the selected abnormal spraying record [D1 i ,D2 i ], if D1 i <D i (min), then one of the feature sets is used as several features to select abnormal spray records. If D2 i >D i (max), then the other set of feature sets is used as several features for selecting abnormal spraying records; the monitoring time point corresponding to any dimensional data is obtained, and several time intervals in which the dimensional data does not belong to the normal value range are extracted to obtain several features of the selected abnormal spraying records under the i-th dimension and the time interval corresponding to each feature; the common time interval is extracted because the influence duration between different features may be different, and the relationship of mutual influence can be better reflected within the same time interval; Step S205: Extract the features of each dimension and the corresponding time interval in all abnormal spraying records respectively, summarize the features extracted from each dimension in the first dimension type set to obtain a first feature set, and summarize the features extracted from each dimension in the second dimension type set to obtain a second feature set.

[0007] Furthermore, step S300 includes the following steps: Step S301: randomly select a spraying record and randomly select two features from the first feature set, and the two features are not in the same feature group; set the dimensions of the two selected features to be the ath dimension and the bth dimension respectively, and obtain the evaluation value of the ath dimension in the selected spraying record as p a and the evaluation value p of the bth dimension b ; Step S302: randomly select a comparison spraying record from the remaining spraying records, and obtain the evaluation value p of the ath dimension in the comparison spraying record. ’ a and the evaluation value p of the bth dimension ’ b ; Get the evaluation difference Δp of the ath dimension between the two spray records respectively a =p a -p ’ a The difference Δp between the evaluation values ​​of the bth dimension b =p b -p ’ b ; Step S303: Obtain the time intervals of the features corresponding to the ath dimension in the two spraying records respectively, take the intersection of the time intervals of the two spraying records to obtain the common time interval of the ath dimension, obtain the data difference of any time point of the ath dimension in the common time interval, and obtain the average difference ΔD1 of the ath dimension in the common time interval a , and the average difference between the bth dimension and the ath dimension in the common time interval is ΔD1 b , then the influence of the ath dimension on the bth dimension is η (a,b) =(ΔD1 a -ΔD1 b ) / ΔD1 a Similarly, obtain the common time interval of the bth dimension in the two spraying records and obtain the influence amplitude η of the bth dimension on the ath dimension (b,a) =(ΔD2 b -ΔD2 a ) / ΔD2 b ; If |η (a,b) |>|η (b,a) |, then the actual influence relationship between the ath dimension and the bth dimension is that the ath dimension affects the bth dimension; the correlation degree g of the ath dimension to the bth dimension is calculated. (a,b) =Δp b / Δp aThe influence relationship between the two dimensions will have a sequence. For example, the spraying distance will affect the spraying range, but the spraying range will have a smaller impact on the spraying distance. Therefore, the actual influence relationship between the two dimensions can be directly reflected, and the magnitude of the influence can well reflect this situation. Step S304: Determine the correlation degree g between the ath dimension and the bth dimension in any two spraying records. (a,b) Obtain and calculate the average value to obtain the correlation degree between the two features G(t1 a ,t1 b ), where t a is the feature belonging to the ath dimension in the first feature set, t b is the feature belonging to the bth dimension in the first feature set.

[0008] Furthermore, step S400 includes the following steps: Step S401: arbitrarily select a feature from the second feature set and set it as the influencing feature, and obtain the dimension of the influencing feature. Arbitrarily select two spraying records, respectively obtain the average value of the dimensional data of the influencing feature in the two spraying records, and obtain the data difference ΔD between the two spraying records of the influencing feature. in ; Set the average value of the dimension data with the larger value in the two spray records as D in (max), and obtain the deviation amplitude f of the influencing feature between the two spraying records D =ΔD in / D in (max); Step S402: arbitrarily select a target feature from the first feature set, obtain the dimension of the target feature, obtain the evaluation difference Δp between the two spraying records in the dimension, set the evaluation value with the larger value in the two spraying records as p(max), and obtain the deviation amplitude f between the two spraying records of the target feature. p =Δp / p(max); Step S403: Obtain the deviation amplitude of any influencing feature between two spraying records from the second feature set, and set the deviation amplitude of the kth influencing feature as (f D ) k , according to the formula: ; Where k is a positive integer and k∈(1,e), e is the number of influencing features contained in the second feature set; the influence weight δ of the kth influencing feature is calculated as k , and get the influence value of the kth influencing feature on the target feature y k =f p×δ k The impact of external environmental factors on the characteristics of each dimension works simultaneously, so the degree of impact can be reflected based on the deviation ratio of each environmental factor. Step S404: Obtain the influence value of the kth influencing feature on the target feature in any two spraying records, calculate the average value to obtain the degree of influence of the kth influencing feature on the target feature; set the target feature as the jth feature in the first feature set, and obtain the degree of influence of the kth influencing feature on the jth feature as Y(k,j).

[0009] Furthermore, step S500 includes the following steps: Step S501: When spraying the aluminum veneer in real time, synchronous monitoring is performed using monitoring equipment, and monitoring data of each monitoring equipment is extracted every unit period to obtain a dimensional data set in each dimension in the unit period; Step S502: Evaluate the dimension data of each dimension according to the evaluation rules to obtain the real-time evaluation value Z at the current moment. now , set the quality assessment threshold to Z th , if Z now <Z th , then an abnormal reminder will be given for the spraying process of the aluminum veneer at the current moment; Step S503: When an abnormality is reported during the spraying process, a dimensional data set of each dimension in the current unit period is obtained, and the dimensional data of each dimension is compared with the corresponding normal value range to obtain a real-time feature set. Each real-time feature in the real-time feature set is then determined to be classified according to whether it belongs to the first feature set; Step S504: Sort the real-time features belonging to the first feature set in ascending order according to the deviation from the normal value range to obtain a feature queue; select the first real-time feature with the smallest deviation according to the queue order, and adjust the dimension of the first real-time feature so that the dimension data of the dimension is within the normal value range, and set the adjustment amplitude to Δf1 now , select any other real-time feature from the feature queue, and obtain the correlation degree G between the first real-time feature and the other real-time features 12 , the expected adjustment range of the remaining real-time features is calculated to be (Δf2 now ) ex =Δf1 now ×(1+G 12 ); Step S505: Obtain the influence of each real-time feature belonging to the second feature set on the remaining real-time features, and set the influence of the vth real-time feature as Y v, according to the formula: ; Wherein, v is a positive integer and v∈(1,u), u is the number of real-time features belonging to the second feature set; the actual adjustment range (Δf2 now ) ac , make corresponding adjustments to the dimensions where the remaining real-time features are located; adjust each real-time feature at the same time, reselect new real-time features for real-time features that are still not within the normal value range after adjustment, and synchronously adjust the remaining real-time features until the dimensional data of the dimensions where all real-time features are located are within the normal value range.

[0010] Aluminum veneer electrostatic spraying intelligent industrial control system, the control system includes historical spraying analysis module, feature extraction analysis module, feature correlation analysis module, feature impact analysis module and real-time spraying adjustment module; The historical spraying analysis module is used to collect multi-dimensional data of each spraying process of aluminum veneer and generate corresponding spraying records; it also conducts multi-dimensional quality assessment on any spraying record and obtains the quality assessment value of any spraying record; The feature extraction and analysis module is used to classify the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; analyze the dimensional data of different categories of spraying records in each dimension to obtain the feature sets of different category dimensions; Feature association analysis module is used to analyze the differences in quality assessment of each feature in the same category in different spraying records, and obtain the degree of association between features of the same category; Feature impact analysis module, used to analyze the deviation of different categories of features in different spraying records to obtain the degree of influence between different categories of features; The real-time spraying adjustment module is used to synchronously collect data whenever the aluminum veneer is sprayed, obtain the real-time evaluation value at the current moment and perform anomaly identification; for any abnormal situation, formulate corresponding adjustment strategies to adjust each abnormal feature.

[0011] Furthermore, the historical spray analysis module includes a spray record collection unit and a coating quality assessment unit; The spraying record acquisition unit is used to collect multi-dimensional data of each spraying process of the aluminum veneer and generate corresponding spraying records; the coating quality assessment unit is used to carry out multi-dimensional quality assessment of any spraying record and obtain the quality assessment value of any spraying record.

[0012] Furthermore, the feature extraction and analysis module includes a dimensional data partitioning unit and a dimensional feature extraction unit; The dimensional data division unit is used to classify the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; the dimensional feature extraction unit is used to extract features of different categories of dimensional data based on the quality assessment differences between different spraying records.

[0013] Furthermore, the real-time spraying adjustment unit includes a real-time spraying acquisition unit and a spraying abnormality adjustment unit; The real-time spraying collection unit is used to synchronously collect data whenever the aluminum veneer is sprayed, obtain the real-time evaluation value at the current moment and perform abnormality identification; the spraying abnormality adjustment unit is used to formulate corresponding adjustment strategies for each abnormal feature according to the existing abnormal situation.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention comprehensively analyzes various factors in the aluminum veneer spraying process to obtain the correlation and influence between different factors. Compared with the isolated factor adjustment in the traditional control system, it can help workers complete the adjustment quickly and greatly improve the accuracy of the adjustment while saving resources. 2. The present invention identifies the influencing factors of the spraying equipment and aluminum veneer during the spraying process and analyzes the correlation between them. When adjusting one of the influencing factors, it will affect the other influencing factors. By understanding this correlation, the abnormal situation can be quickly adjusted, greatly improving the processing efficiency. 3. The present invention also analyzes the external environmental factors that affect the spraying process, captures the degree of influence of each environmental factor, and can take into account the influence of the environment when adjusting parameters, adjust relevant parameters more accurately, and help staff improve the accuracy of the adjustment process, so as to effectively achieve the desired effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the steps of an intelligent industrial control method for electrostatic spraying of aluminum veneer; Figure 2 This is a structural diagram of the intelligent industrial control system for electrostatic spraying of aluminum veneer. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0017] Example: Figures 1 to 2As shown, the present invention provides an intelligent industrial control method for electrostatic spraying of aluminum veneer, and the control method includes the following steps: Step S100: performing multi-dimensional data collection on each spraying process of the aluminum veneer to generate corresponding spraying records; performing multi-dimensional quality assessment on any spraying record to obtain a quality assessment value of any spraying record; Wherein, step S100 includes the following steps: Step S101: Several monitoring devices are installed to continuously monitor and collect data during the spraying process of the aluminum veneer, and the data type of the monitoring data is obtained and used as the dimension of the monitoring device; all monitoring data collected by any monitoring device are obtained to obtain a dimensional data set of the corresponding dimension, and the dimensional data set of each monitoring device is summarized to generate a spraying record; Step S102: randomly select a spraying record, obtain the dimension data of each dimension in the spraying record, preset corresponding evaluation rules for each dimension, evaluate the dimension data of each dimension to obtain the evaluation value of the dimension; preset an influence weight for each dimension, set the evaluation value of the i-th dimension in the spraying record to p i and the influence weight is μ i , according to the formula: ; Wherein, i is a positive integer and i∈(1,a), a is the number of dimensions in the spraying record; the quality evaluation value Z of the spraying record is calculated.

[0018] Step S200: classifying the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; analyzing the dimensional data of different categories of spraying records in each dimension to obtain feature sets in different categories and dimensions; Step S200 includes the following steps: Step S201: Pre-build an equipment dimension database, which contains several dimension categories. Select any one dimension from the several dimensions in the spraying record, and compare the selected dimension with each dimension category to obtain the similarity between the selected dimension and each dimension category; select the dimension category with the greatest similarity and set the maximum similarity as S. max , preset a similarity threshold S th , if S max ≥S th , then the selected dimension is set as the first dimension type set. If S max <S th , then the selected dimension is set as the second dimension type set; Step S202: Preset a quality assessment threshold Z th, if there is a spray record that satisfies Z<Z th , then the corresponding spray record is set as an abnormal spray record. If Z≥Z th , it is set as a normal spraying record; arbitrarily select a normal spraying record, and arbitrarily select the dimension data set of the i-th dimension from the selected normal spraying record. If the i-th dimension belongs to the first dimension type set, the value range of the dimension data set is obtained; obtain the value range of the i-th dimension in each normal spraying record and perform a union operation to obtain the normal value range of the i-th dimension [D i (min),D i (max)]; Step S203: Set two sets of feature sets for the i-th dimension, and set the dimension data of the i-th dimension to D i , when D is satisfied i <D i (min), then one of the feature sets is selected as the dimension data D i The feature set of i >D i (max), another set of feature sets is selected as the dimension data D i The feature set of Step S204: randomly select an abnormal spraying record and obtain the value range of the i-th dimension in the selected abnormal spraying record [D1 i ,D2 i ], if D1 i <D i (min), then one of the feature sets is used as several features to select abnormal spray records. If D2 i >D i (max), then the other set of feature sets is used as several features for selecting abnormal spray records; obtaining the monitoring time point corresponding to any dimensional data, extracting several time intervals in which the dimensional data does not belong to the normal value range, and obtaining several features of the selected abnormal spray records under the i-th dimension and the time interval corresponding to each feature; Step S205: Extract the features of each dimension and the corresponding time interval in all abnormal spraying records respectively, summarize the features extracted from each dimension in the first dimension type set to obtain a first feature set, and summarize the features extracted from each dimension in the second dimension type set to obtain a second feature set.

[0019] Step S300: Analyze the differences in quality assessment of each feature in the same category in different spraying records to obtain the correlation degree between features in the same category; Wherein, step S300 includes the following steps: Step S301: randomly select a spraying record and randomly select two features from the first feature set, and the two features are not in the same feature group; set the dimensions of the two selected features to be the ath dimension and the bth dimension respectively, and obtain the evaluation value of the ath dimension in the selected spraying record as p a and the evaluation value p of the bth dimension b ; Step S302: randomly select a comparison spraying record from the remaining spraying records, and obtain the evaluation value p of the ath dimension in the comparison spraying record. ’ a and the evaluation value p of the bth dimension ’ b ; Get the evaluation difference Δp of the ath dimension between the two spray records respectively a =p a -p ’ a The difference Δp between the evaluation values ​​of the bth dimension b =p b -p ’ b ; Step S303: Obtain the time intervals of the features corresponding to the ath dimension in the two spraying records respectively, take the intersection of the time intervals of the two spraying records to obtain the common time interval of the ath dimension, obtain the data difference of any time point of the ath dimension in the common time interval, and obtain the average difference ΔD1 of the ath dimension in the common time interval a , and the average difference between the bth dimension and the ath dimension in the common time interval is ΔD1 b , then the influence of the ath dimension on the bth dimension is η (a,b) =(ΔD1 a -ΔD1 b ) / ΔD1 a Similarly, obtain the common time interval of the bth dimension in the two spraying records and obtain the influence amplitude η of the bth dimension on the ath dimension (b,a) =(ΔD2 b -ΔD2 a ) / ΔD2 b ; If |η (a,b) |>|η (b,a) |, then the actual influence relationship between the ath dimension and the bth dimension is that the ath dimension affects the bth dimension; the correlation degree g of the ath dimension to the bth dimension is calculated. (a,b) =Δp b / Δp a ; Example 1: arbitrarily select a dimension a in the common time interval between two spraying records, and obtain the average difference of the selected dimension in the common time interval as 5. Similarly, obtain the average difference of another dimension b in the common time interval of dimension a as 4, and obtain the influence of a on b as 1 / 5=0.2; obtain the average difference of dimension a in the common time interval of dimension b as 4, and the average difference of dimension b in its own common time interval as 2, and obtain the influence of dimension b on dimension a as 2 / 4=0.5. Therefore, 0.5>0.2, and the influence relationship is that dimension b affects dimension a. Step S304: Determine the correlation degree g between the ath dimension and the bth dimension in any two spraying records. (a,b) Obtain and calculate the average value to obtain the correlation degree between the two features G(t1 a ,t1 b ), where t a is the feature belonging to the ath dimension in the first feature set, t b is the feature belonging to the bth dimension in the first feature set.

[0020] Step S400: Analyze the deviation of different types of features in different spraying records to obtain the degree of influence between different types of features; Step S400 includes the following steps: Step S401: arbitrarily select a feature from the second feature set and set it as the influencing feature, and obtain the dimension of the influencing feature. Arbitrarily select two spraying records, respectively obtain the average value of the dimensional data of the influencing feature in the two spraying records, and obtain the data difference ΔD between the two spraying records of the influencing feature. in ; Set the average value of the dimension data with the larger value in the two spray records as D in (max), and obtain the deviation amplitude f of the influencing feature between the two spraying records D =ΔD in / D in (max); Step S402: arbitrarily select a target feature from the first feature set, obtain the dimension of the target feature, obtain the evaluation difference Δp between the two spraying records in the dimension, set the evaluation value with the larger value in the two spraying records as p(max), and obtain the deviation amplitude f between the two spraying records of the target feature. p =Δp / p(max); Step S403: Obtain the deviation amplitude of any influencing feature between two spraying records from the second feature set, and set the deviation amplitude of the kth influencing feature as (f D ) k , according to the formula: ; Where k is a positive integer and k∈(1,e), e is the number of influencing features contained in the second feature set; the influence weight δ of the kth influencing feature is calculated as k , and get the influence value of the kth influencing feature on the target feature y k =f p ×δ k ; Step S404: Obtain the influence value of the kth influencing feature on the target feature in any two spraying records, calculate the average value to obtain the degree of influence of the kth influencing feature on the target feature; set the target feature as the jth feature in the first feature set, and obtain the degree of influence of the kth influencing feature on the jth feature as Y(k,j).

[0021] Step S500: Whenever the aluminum veneer is sprayed, data is collected synchronously to obtain the real-time evaluation value at the current moment and perform abnormality identification; for any abnormal situation, a corresponding adjustment strategy is formulated to adjust each abnormal feature; Step S500 includes the following steps: Step S501: When spraying the aluminum veneer in real time, synchronous monitoring is performed using monitoring equipment, and monitoring data of each monitoring equipment is extracted every unit period to obtain a dimensional data set in each dimension in the unit period; Step S502: Evaluate the dimension data of each dimension according to the evaluation rules to obtain the real-time evaluation value Z at the current moment. now , set the quality assessment threshold to Z th , if Z now <Z th , then an abnormal reminder will be given for the spraying process of the aluminum veneer at the current moment; Step S503: When an abnormality is reported during the spraying process, a dimensional data set of each dimension in the current unit period is obtained, and the dimensional data of each dimension is compared with the corresponding normal value range to obtain a real-time feature set. Each real-time feature in the real-time feature set is then determined to be classified according to whether it belongs to the first feature set; Step S504: Sort the real-time features belonging to the first feature set in ascending order according to the deviation from the normal value range to obtain a feature queue; select the first real-time feature with the smallest deviation according to the queue order, and adjust the dimension of the first real-time feature so that the dimension data of the dimension is within the normal value range, and set the adjustment amplitude to Δf1 now, select any other real-time feature from the feature queue, and obtain the correlation degree G between the first real-time feature and the other real-time features 12 , the expected adjustment range of the remaining real-time features is calculated to be (Δf2 now ) ex =Δf1 now ×(1+G 12 ); Step S505: Obtain the influence of each real-time feature belonging to the second feature set on the remaining real-time features, and set the influence of the vth real-time feature as Y v , according to the formula: ; Wherein, v is a positive integer and v∈(1,u), u is the number of real-time features belonging to the second feature set; the actual adjustment range (Δf2 now ) ac , make corresponding adjustments to the dimensions of the remaining real-time features; adjust each real-time feature simultaneously, reselect new real-time features for adjustments to real-time features that are still not within the normal value range after adjustment, and synchronously adjust the remaining real-time features until the dimensional data of the dimensions of all real-time features are within the normal value range; Example 2: Set the adjustment range of dimension a to 5, obtain the correlation between dimension b and dimension a as 0.2, and obtain the expected adjustment range of dimension b as 5×(1+0.2)=6; obtain three real-time features belonging to the second feature set, with influence levels of 1%, 2%, and 3%, respectively, and calculate the actual adjustment range of dimension b as 5×(1+0.01)×(1+0.02)×(1+0.03)=5.30553; Aluminum veneer electrostatic spraying intelligent industrial control system, the control system includes historical spraying analysis module, feature extraction analysis module, feature correlation analysis module, feature impact analysis module and real-time spraying adjustment module; The historical spraying analysis module is used to collect multi-dimensional data of each spraying process of aluminum veneer and generate corresponding spraying records; it also conducts multi-dimensional quality assessment on any spraying record and obtains the quality assessment value of any spraying record; The feature extraction and analysis module is used to classify the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; analyze the dimensional data of different categories of spraying records in each dimension to obtain the feature sets of different category dimensions; Feature association analysis module is used to analyze the differences in quality assessment of each feature in the same category in different spraying records, and obtain the degree of association between features of the same category; Feature impact analysis module, used to analyze the deviation of different categories of features in different spraying records to obtain the degree of influence between different categories of features; The real-time spraying adjustment module is used to synchronously collect data whenever the aluminum veneer is sprayed, obtain the real-time evaluation value at the current moment and perform anomaly identification; for any abnormal situation, formulate corresponding adjustment strategies to adjust each abnormal feature.

[0022] Among them, the historical spray analysis module includes a spray record collection unit and a coating quality assessment unit; The spraying record acquisition unit is used to collect multi-dimensional data of each spraying process of the aluminum veneer and generate corresponding spraying records; the coating quality assessment unit is used to carry out multi-dimensional quality assessment of any spraying record and obtain the quality assessment value of any spraying record.

[0023] Among them, the feature extraction and analysis module includes a dimension data division unit and a dimension feature extraction unit; The dimensional data division unit is used to classify the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; the dimensional feature extraction unit is used to extract features of different categories of dimensional data based on the quality assessment differences between different spraying records.

[0024] Among them, the real-time spraying adjustment unit includes a real-time spraying acquisition unit and a spraying abnormality adjustment unit; The real-time spraying collection unit is used to synchronously collect data whenever the aluminum veneer is sprayed, obtain the real-time evaluation value at the current moment and perform abnormality identification; the spraying abnormality adjustment unit is used to formulate corresponding adjustment strategies for each abnormal feature according to the existing abnormal situation.

[0025] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. Intelligent industrial control method for electrostatic spraying of aluminum veneer, characterized by: The control method comprises the following steps: Step S100: performing multi-dimensional data collection on each spraying process of the aluminum veneer to generate corresponding spraying records; performing multi-dimensional quality assessment on any spraying record to obtain a quality assessment value of any spraying record; Step S200: classifying the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; analyzing the dimensional data of different categories of spraying records in each dimension to obtain feature sets in different categories and dimensions; Step S300: Analyze the differences in quality assessment of each feature in the same category in different spraying records to obtain the correlation degree between features in the same category; Step S400: Analyze the deviation of different types of features in different spraying records to obtain the degree of influence between different types of features; Step S500: Whenever the aluminum veneer is sprayed, data is collected synchronously to obtain the real-time evaluation value at the current moment and perform abnormality identification; for any abnormal situation, a corresponding adjustment strategy is formulated to adjust each abnormal feature.

2. The intelligent industrial control method for electrostatic spraying of aluminum veneer according to claim 1 is characterized in that: The step S100 includes the following steps: Step S101: Several monitoring devices are installed to continuously monitor and collect data during the spraying process of the aluminum veneer, and the data type of the monitoring data is obtained and used as the dimension of the monitoring device; all monitoring data collected by any monitoring device are obtained to obtain a dimensional data set of the corresponding dimension, and the dimensional data set of each monitoring device is summarized to generate a spraying record; Step S102: randomly select a spraying record, obtain the dimension data of each dimension in the spraying record, preset corresponding evaluation rules for each dimension, evaluate the dimension data of each dimension to obtain the evaluation value of the dimension; preset an influence weight for each dimension, set the evaluation value of the i-th dimension in the spraying record to p i and the influence weight is μ i , according to the formula: ； Wherein, i is a positive integer and i∈(1,a), a is the number of dimensions in the spraying record; the quality evaluation value Z of the spraying record is calculated.

3. The intelligent industrial control method for electrostatic spraying of aluminum veneer according to claim 2 is characterized in that: The step S200 includes the following steps: Step S201: Pre-build an equipment dimension database, which contains several dimension categories. Select any one dimension from the several dimensions in the spraying record, and compare the selected dimension with each dimension category to obtain the similarity between the selected dimension and each dimension category; select the dimension category with the greatest similarity and set the maximum similarity as S. max , preset a similarity threshold S th , if S max ≥S th , then the selected dimension is set as the first dimension type set. If S max <S th , then the selected dimension is set as the second dimension type set; Step S202: Preset a quality assessment threshold Z th , if there is a spray record that satisfies Z<Z th , then the corresponding spray record is set as an abnormal spray record. If Z≥Z th , it is set as a normal spraying record; arbitrarily select a normal spraying record, and arbitrarily select the dimension data set of the i-th dimension from the selected normal spraying record. If the i-th dimension belongs to the first dimension type set, the value range of the dimension data set is obtained; obtain the value range of the i-th dimension in each normal spraying record and perform a union operation to obtain the normal value range of the i-th dimension [D i (min),D i (max)]; Step S203: Set two sets of feature sets for the i-th dimension, and set the dimension data of the i-th dimension to D i , when D is satisfied i <D i (min), then one of the feature sets is selected as the dimension data D i The feature set of i >D i (max), another set of feature sets is selected as the dimension data D i The feature set of Step S204: randomly select an abnormal spraying record and obtain the value range of the i-th dimension in the selected abnormal spraying record [D1 i ,D2 i ], if D1 i <D i (min), then one of the feature sets is used as several features to select abnormal spray records. If D2 i >D i (max), then the other set of feature sets is used as several features for selecting abnormal spray records; obtaining the monitoring time point corresponding to any dimensional data, extracting several time intervals in which the dimensional data does not belong to the normal value range, and obtaining several features of the selected abnormal spray records under the i-th dimension and the time interval corresponding to each feature; Step S205: Extract the features of each dimension and the corresponding time interval in all abnormal spraying records respectively, summarize the features extracted from each dimension in the first dimension type set to obtain a first feature set, and summarize the features extracted from each dimension in the second dimension type set to obtain a second feature set.

4. The intelligent industrial control method for electrostatic spraying of aluminum veneer according to claim 3 is characterized in that: The step S300 includes the following steps: Step S301: randomly select a spraying record and randomly select two features from the first feature set, and the two features are not in the same feature group; set the dimensions of the two selected features to be the ath dimension and the bth dimension respectively, and obtain the evaluation value of the ath dimension in the selected spraying record as p a and the evaluation value p of the bth dimension b ; Step S302: randomly select a comparison spraying record from the remaining spraying records, and obtain the evaluation value p of the ath dimension in the comparison spraying record. ’ a and the evaluation value p of the bth dimension ’ b ; Get the evaluation difference Δp of the ath dimension between the two spray records respectively a =p a -p ’ a The difference Δp between the evaluation values ​​of the bth dimension b =p b -p ’ b ; Step S303: Obtain the time intervals of the features corresponding to the ath dimension in the two spraying records respectively, take the intersection of the time intervals of the two spraying records to obtain the common time interval of the ath dimension, obtain the data difference of any time point of the ath dimension in the common time interval, and obtain the average difference ΔD1 of the ath dimension in the common time interval a , and the average difference between the bth dimension and the ath dimension in the common time interval is ΔD1 b , then the influence of the ath dimension on the bth dimension is η (a,b) =(ΔD1 a -ΔD1 b ) / ΔD1 a Similarly, obtain the common time interval of the bth dimension in the two spraying records and obtain the influence amplitude η of the bth dimension on the ath dimension (b,a) =(ΔD2 b -ΔD2 a ) / ΔD2 b ; If |η (a,b) |>|η (b,a) |, then the actual influence relationship between the ath dimension and the bth dimension is that the ath dimension affects the bth dimension; the correlation degree g of the ath dimension to the bth dimension is calculated. (a,b) =Δp b / Δp a ; Step S304: Determine the correlation degree g between the ath dimension and the bth dimension in any two spraying records. (a,b) Obtain and calculate the average value to obtain the correlation degree between the two features G(t1 a ,t1 b ), where t a is the feature belonging to the ath dimension in the first feature set, t b is the feature belonging to the bth dimension in the first feature set.

5. The intelligent industrial control method for electrostatic spraying of aluminum veneer according to claim 4 is characterized in that: The step S400 includes the following steps: Step S401: arbitrarily select a feature from the second feature set and set it as the influencing feature, and obtain the dimension of the influencing feature. Arbitrarily select two spraying records, respectively obtain the average value of the dimensional data of the influencing feature in the two spraying records, and obtain the data difference ΔD between the two spraying records of the influencing feature. in ; Set the average value of the dimension data with the larger value in the two spray records as D in (max), and obtain the deviation amplitude f of the influencing feature between the two spraying records D =ΔD in / D in (max); Step S402: arbitrarily select a target feature from the first feature set, obtain the dimension of the target feature, obtain the evaluation difference Δp between the two spraying records in the dimension, set the evaluation value with the larger value in the two spraying records as p(max), and obtain the deviation amplitude f between the two spraying records of the target feature. p =Δp / p(max); Step S403: Obtain the deviation amplitude of any influencing feature between two spraying records from the second feature set, and set the deviation amplitude of the kth influencing feature as (f D ) k , according to the formula: ； Where k is a positive integer and k∈(1,e), e is the number of influencing features contained in the second feature set; the influence weight δ of the kth influencing feature is calculated as k , and get the influence value of the kth influencing feature on the target feature y k =f p ×δ k ; Step S404: Obtain the influence value of the kth influencing feature on the target feature in any two spraying records, calculate the average value to obtain the degree of influence of the kth influencing feature on the target feature; set the target feature as the jth feature in the first feature set, and obtain the degree of influence of the kth influencing feature on the jth feature as Y(k,j).

6. The intelligent industrial control method for electrostatic spraying of aluminum veneer according to claim 5 is characterized in that: The step S500 includes the following steps: Step S501: When spraying the aluminum veneer in real time, synchronous monitoring is performed using monitoring equipment, and monitoring data of each monitoring equipment is extracted every unit period to obtain a dimensional data set in each dimension in the unit period; Step S502: Evaluate the dimension data of each dimension according to the evaluation rules to obtain the real-time evaluation value Z at the current moment. now , set the quality assessment threshold to Z th , if Z now <Z th , then an abnormal reminder will be given for the spraying process of the aluminum veneer at the current moment; Step S503: When an abnormality is reported during the spraying process, a dimensional data set of each dimension in the current unit period is obtained, and the dimensional data of each dimension is compared with the corresponding normal value range to obtain a real-time feature set. Each real-time feature in the real-time feature set is then determined to be classified according to whether it belongs to the first feature set; Step S504: Sort the real-time features belonging to the first feature set in ascending order according to the deviation from the normal value range to obtain a feature queue; select the first real-time feature with the smallest deviation according to the queue order, and adjust the dimension of the first real-time feature so that the dimension data of the dimension is within the normal value range, and set the adjustment amplitude to Δf1 now , select any other real-time feature from the feature queue, and obtain the correlation degree G between the first real-time feature and the other real-time features 12 , the expected adjustment range of the remaining real-time features is calculated as (Δf2 now ) ex =Δf1 now ×(1+G 12 ); Step S505: Obtain the influence of each real-time feature belonging to the second feature set on the remaining real-time features, and set the influence of the vth real-time feature as Y v , according to the formula: ； Wherein, v is a positive integer and v∈(1,u), u is the number of real-time features belonging to the second feature set; the actual adjustment range (Δf2 now ) ac , make corresponding adjustments to the dimensions where the remaining real-time features are located; adjust each real-time feature at the same time, reselect new real-time features for real-time features that are still not within the normal value range after adjustment, and synchronously adjust the remaining real-time features until the dimensional data of the dimensions where all real-time features are located are within the normal value range.

7. An intelligent industrial control system for electrostatic spraying of aluminum veneers, configured to execute the intelligent industrial control method for electrostatic spraying of aluminum veneers according to any one of claims 1 to 6, characterized in that: The control system includes a historical spray analysis module, a feature extraction analysis module, a feature correlation analysis module, a feature impact analysis module and a real-time spray adjustment module; The historical spraying analysis module is used to collect multi-dimensional data of each spraying process of aluminum veneer and generate corresponding spraying records; carry out multi-dimensional quality assessment on any spraying record and obtain the quality assessment value of any spraying record; The feature extraction and analysis module is used to classify the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; analyze the dimensional data of different categories of spraying records in each dimension to obtain feature sets in different categories and dimensions respectively; The feature association analysis module is used to analyze the differences in quality assessment of each feature in the same category in different spraying records, and obtain the degree of association between features in the same category; The feature impact analysis module is used to analyze the deviation of different categories of features in different spraying records to obtain the degree of influence between different categories of features; The real-time spraying adjustment module is used to synchronously collect data whenever the aluminum veneer is sprayed, obtain the real-time evaluation value at the current moment and perform abnormality identification; for any abnormal situation that exists, formulate corresponding adjustment strategies to adjust each abnormal feature.

8. The intelligent industrial control system for electrostatic spraying of aluminum veneer according to claim 7 is characterized by: The historical spray analysis module includes a spray record collection unit and a coating quality assessment unit; The spraying record acquisition unit is used to collect multi-dimensional data of each spraying process of the aluminum veneer and generate corresponding spraying records; the coating quality assessment unit is used to carry out multi-dimensional quality assessment of any spraying record and obtain the quality assessment value of any spraying record.

9. The intelligent industrial control system for electrostatic spraying of aluminum veneer according to claim 7, characterized in that: The feature extraction and analysis module includes a dimensional data division unit and a dimensional feature extraction unit; The dimensional data division unit is used to classify the dimensional data of each dimension according to the data source of the dimensional data in different dimensions; the dimensional feature extraction unit is used to extract features of different categories of dimensional data based on the quality assessment differences between different spraying records.

10. The intelligent industrial control system for electrostatic spraying of aluminum veneer according to claim 7, characterized in that: The real-time spraying adjustment unit includes a real-time spraying acquisition unit and a spraying abnormality adjustment unit; The real-time spraying acquisition unit is used to synchronously collect data whenever the aluminum veneer is sprayed, obtain the real-time evaluation value at the current moment and perform abnormality identification; the spraying abnormality adjustment unit is used to formulate corresponding adjustment strategies for each abnormal feature according to the existing abnormal situation.

Images