Adjustment management system based on PLC control

Abstract

The invention belongs to the technical field of automation control, and discloses an adjustment management system based on PLC control. Comprising a data acquisition module used for acquiring industrial state data in real time; the data detection module is used for performing anomaly detection on the industrial state data and re-collecting abnormal data; the state evaluation module is used for evaluating the industrial state data and judging whether an adjusting instruction is generated or not; the data optimization module is used for marking the industrial state data corresponding to the adjustment instruction as outlier data and dynamically optimizing the outlier data; the control adjustment module is used for performing parameter adjustment on the actuator corresponding to the outlier data; the monitoring and adjusting capability in the industrial production process can be effectively improved, and it is ensured that industrial equipment is in the optimal operation state; the capability of responding to potential faults and abnormal conditions is enhanced, so that potential safety hazards are reduced; and the industrial production process is promoted to develop towards more automatic and intelligent directions.

Description

Technical Field

[0001] The present invention relates to the field of automation control technology, and more specifically, to a regulation management system based on PLC control. Background Art

[0002] In modern industrial production, the rapid development of automation technology has promoted the improvement of production efficiency and product quality; in particular, the widespread application of programmable logic controllers (PLCs) has made the monitoring and regulation of various production processes more flexible and precise; PLC control systems can respond to changes in real time in complex and dynamic environments, becoming a core component in the field of industrial automation; in this context, PLC-based regulation management systems have emerged, which can realize comprehensive monitoring, data collection and intelligent regulation of the production process, ensuring that the industrial production process runs in the best state; it not only improves production efficiency and quality, but also promotes industrial automation to develop in a more intelligent direction.

[0003] The patent application with publication number CN118092302A discloses an automatic metering device and control method for acid loading in an acid truck; comprising: a metering valve body system, the metering valve body system comprises a flow meter, a disc valve, an electric switch valve and a regulating valve connected in series in sequence; a PLC control system, the PLC control system at least comprises an acid release amount accumulation unit and a regulating valve opening adjustment unit, the acid release amount accumulation unit is used to receive a flow meter sensor signal and determine the accumulated acid release amount, the regulating valve opening adjustment unit is used to output a regulating valve opening adjustment signal to the regulating valve based on the difference between the accumulated acid release amount and the given acid release amount; this invention combines the comparison of the acid loading amount with the given loading amount and the size of the flow rate to control the opening of the regulating valve, which can effectively ensure the safe production, scheduling, management and optimization of the acid loading process of the acid truck, ensure the accuracy of the acid loading, ensure that the acid loading amount is not overloaded, and reduce the risk and safety hazards when the acid release valve fails.

[0004] However, although the above technology realizes the regulation management of the control valve based on the PLC control system, the opening of the control valve is controlled by a PID controller; the parameters in the PID controller (such as proportional, integral, differential gain, etc.) usually need to be adjusted manually, which is difficult to adapt to the dynamically changing system characteristics. In a highly nonlinear environment, the PID controller is prone to overshoot or instability; in addition, the PID controller has poor robustness to system parameter changes and external disturbances, and is prone to large control errors; therefore, if the above technology is derived into industrial production, it will lead to reduced production efficiency, unstable production quality and increased safety hazards.

[0005] In view of this, the present invention proposes a regulation management system based on PLC control to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a regulation management system based on PLC control, comprising:

[0007] Data acquisition module, used to collect industrial status data in real time;

[0008] Data detection module, used to detect anomalies in industrial status data and re-collect abnormal data;

[0009] A status evaluation module is used to evaluate industrial status data and determine whether to generate adjustment instructions;

[0010] A data optimization module is used to mark the industrial status data corresponding to the adjustment instruction as outlier data and dynamically optimize the outlier data;

[0011] The control and adjustment module is used to adjust the parameters of the actuator corresponding to the outlier data.

[0012] Furthermore, the industrial status data includes equipment operation data, process data and production environment data;

[0013] The step of performing anomaly detection on industrial status data comprises:

[0014] Step S101: setting a digital label for each data in the industrial status data in turn and marking it as a data label, the range of the data label is [1, θ], θ is the number of data in the industrial status data;

[0015] Step S102: Select data with data label υ for anomaly detection, υ∈[1,θ];

[0016] Step S103: let υ=υ+1;

[0017] Step S104: looping steps S102 to S103 until all data in the industrial status data have completed abnormality detection, and the loop ends;

[0018] In step S102, the method for abnormality detection includes:

[0019] Mark the data for anomaly detection as elements; obtain a historical elements, where the historical elements are elements collected at historical moments, and a is an integer greater than 1; preset the analysis quantity b, mark the elements collected in real time as real-time elements, and use the real-time elements and historical elements as analysis data; subtract the corresponding neighbor elements from each element in the analysis data to obtain the element difference corresponding to each element; wherein the neighbor elements corresponding to the element are the remaining a elements in the analysis data; determine the reference distance and standard element corresponding to each element based on the analysis quantity and element difference; determine the adjacent distance between each element and each corresponding standard element based on the reference distance and element difference of each element; calculate the local density of each element based on the adjacent distance between each element and each corresponding standard element; calculate the anomaly coefficient corresponding to the real-time element based on the adjacent distance between each element and each corresponding standard element;

[0020] Preset the quantity range [b, b+c], set the analysis quantity to each value in the quantity range in turn, and calculate the abnormal coefficient corresponding to each value in the quantity range; add all the abnormal coefficients in turn, and then divide them by c+1 to obtain the abnormal mean; preset the judgment coefficient d, 0<d<1; if |yz-1|≥d, the data corresponding to the real-time element is marked as abnormal data, yz is the abnormal mean; if |yz-1|<d, the data corresponding to the real-time element is not marked.

[0021] Furthermore, the method for determining the reference distance and the standard element corresponding to each element includes:

[0022] All element differences corresponding to each element are taken as an element set, and the element set corresponds to the element one by one; the element differences in each element set are sorted from small to large to generate an element sorting table corresponding to each element set; the element difference ranked in the bth position in each element sorting table is taken as the reference distance of the element corresponding to the corresponding element set, and the neighbor element corresponding to the first b element differences in each element sorting table is taken as the standard element of the element corresponding to the corresponding element set;

[0023] The expression of the adjacent distance is: xl(p,q)=max(jl(p),yc(p,q)); where xl(p,q) is the adjacent distance between the p-th element in the analysis data and the corresponding q-th standard element, max is the maximum function, jl(p) is the reference distance of the p-th element in the analysis data, yc(p,q) is the element difference between the p-th element in the analysis data and the corresponding q-th standard element, p∈[1,a+1], q∈[1,b], a>b;

[0024] The expression of local density is: Where md(p) is the local density of the pth element in the analysis data;

[0025] The expression of the anomaly coefficient is: Where yx is the anomaly coefficient corresponding to the real-time element, md′ is the local density corresponding to the real-time element, and md(q) is the local density of the qth canonical element corresponding to the real-time element.

[0026] Furthermore, the step of determining whether to generate an adjustment instruction includes:

[0027] Step S201: Select data with data label υ for outlier detection;

[0028] Step S202: let υ=υ+1;

[0029] Step S203: looping steps S201 to S202 until all data in the industrial status data have completed outlier detection, the loop ends, and the process proceeds to step S204;

[0030] Step S204: determining whether there is data in the industrial status data that is marked as outlier data, if so, generating an adjustment instruction, if not, not generating an adjustment instruction;

[0031] In step S201, the outlier detection method includes:

[0032] Mark the data for outlier detection as nodes, and mark the nodes collected in real time as real-time nodes; obtain g historical nodes, where the historical nodes are normal nodes collected at historical moments, the normal nodes are nodes within the normal range, and g is an integer greater than 1; add the g historical nodes in sequence, and then divide by g to obtain the mean vector; calculate the covariance matrix corresponding to the historical nodes based on the mean vector; calculate the covariance distance corresponding to the real-time nodes based on the covariance matrix;

[0033] Set the significance level; obtain the data dimension corresponding to the node, where the data dimension is the number of data types in the node; obtain the distance threshold based on the significance level and data dimension; square the covariance distance to obtain the square distance, and compare the square distance with the distance threshold; if the square distance is greater than the distance threshold, mark the data corresponding to the corresponding node as outlier data; if the square distance is less than or equal to the distance threshold, do not mark the data corresponding to the corresponding node.

[0034] Furthermore, the expression of the covariance matrix is: Where Σ is the covariance matrix, C h is the hth node, (C h -μ) T For (C h -μ), h∈[1,g];

[0035] The expression of covariance distance is: Where D is the covariance distance, A real-time node.

[0036] Furthermore, the step of dynamically optimizing the outlier data includes:

[0037] Step S301: Obtain the normal range corresponding to each outlier data;

[0038] Step S302: randomly selecting a value from the normal range corresponding to each outlier data to construct a set of data sets, constructing a total of M sets of data sets, and the M sets of data sets are all different; setting a different digital label for each set of data sets and marking it as a set label;

[0039] Step S303: construct a wave group S, which includes m water waves, the position of each water wave corresponds to the set label one by one, and set the number of iterations t to 0;

[0040] Step S304: defining a performance indicator function and an iteration threshold T;

[0041] Step S305: Divide the wave group S into u sub-wave groups;

[0042] Step S306: propagate each water wave and update the position of each water wave;

[0043] Step S307: Calculate the reflection probability of each water wave and update the position of the reflecting water wave;

[0044] Step S308: interfering each water wave and updating the position of each water wave;

[0045] Step S309: compare the number of iterations t with the iteration threshold T; if t≥T, proceed to step S310; if t<T, set t=t+1 and return to step S306;

[0046] Step S310: Calculate the performance index of each water wave, obtain the water wave with the largest performance index, and mark it as the best water wave; dynamically optimize the outlier data according to the data set corresponding to the set label corresponding to the best water wave.

[0047] Furthermore, in step S301, the method for obtaining the normal range corresponding to each outlier data includes:

[0048] The historical nodes corresponding to each outlier data are taken as a set of historical sets, and the historical sets correspond to the outlier data one by one; the historical node with the largest value in each historical set is taken as the maximum value of the corresponding outlier data; the historical node with the smallest value in each historical set is taken as the minimum value of the corresponding outlier data; based on the maximum and minimum values ​​of each outlier data, the normal range corresponding to each outlier data is constructed;

[0049] In step S303, the position of each water wave in the wave group S is defined in a one-dimensional search space, and the range of the one-dimensional search space is the set label corresponding to the M data sets; the expression of the position of each water wave in the wave group S is: In the formula, is the position of the i-th water wave, P i is the random coefficient of the ith water wave, P i ∈[0, 1], i∈[1,m];

[0050] In the step S304, the expression of the performance index function is: f=ω1×sx+ω2×sz; wherein f is the performance index, sx is the production efficiency, sz is the production quality, ω1 and ω2 are preset weight coefficients; the method for obtaining the production efficiency and production quality is: obtaining the data set corresponding to the set label corresponding to the water wave position, and marking it as a replacement set; replacing all outlier data in the industrial status data with corresponding values ​​in the replacement set, and marking the replaced industrial status data as predicted data; inputting the predicted data into the trained efficiency prediction model to predict the corresponding production efficiency; inputting the predicted data into the trained quality prediction model to predict the corresponding production quality; the training processes of the efficiency prediction model and the quality prediction model are consistent, and both are deep neural network models.

[0051] Furthermore, in step S305, the method of dividing the wave group S into u sub-wave groups includes:

[0052] Calculate the performance index corresponding to each water wave in the wave group S, and sort them from large to small; set an increasing serial number for each water wave in the positive order of sorting, and the serial number range is [1, m]; according to u sub-wave groups, perform a modulo operation on the serial number of each water wave to obtain the corresponding sub-sequence number; the expression of the sub-sequence number is: zx = xl%u; where zx is the sub-sequence number, xl is the sequence number, and % is the modulo function; if the sub-sequence number is not 0, the corresponding water wave is assigned to the zxth sub-wave group; if the sub-sequence number is 0, the corresponding water wave is assigned to the uth sub-wave group; each sub-wave group includes k water waves, and m = uk;

[0053] In step S306, the method for updating the position of each water wave includes:

[0054]

[0055] In the formula, is the position of the i-th water wave after propagation, is the position of the i-th water wave before propagation, is the position of the local optimal water wave, X b is the position of the global optimal water wave, is the position of the wave with the largest performance index in the sub-wave group corresponding to the i-th wave, and E1 and E2 are both random numbers between [0,1]. The local optimal wave is the wave with the largest performance index in this iteration, and the global optimal wave is the wave with the largest performance index in all iterations.

[0056] Furthermore, in step S307, the expression of reflection probability is: In the formula, is the reflection probability of the i-th water wave, e is a natural constant, f i t is the performance index corresponding to the i-th water wave, f b is the performance index corresponding to the global optimal water wave;

[0057] For each water wave, a random number L between [0,1] is generated i ,like Then the corresponding water wave is marked as reflected water wave. The corresponding water wave is not marked, and the water wave marked as the reflected water wave is reflected; the method for updating the position of the reflected water wave includes:

[0058]

[0059] In the formula, is the position of the nth water wave after reflection, is the position of the nth water wave before reflection, y is the reflection factor, l is the propagation factor, is the performance index corresponding to the nth water wave being reflected, n∈[1,N], N is the number of reflected water waves, and r is the gradient factor;

[0060] In step S308, the method for updating the position of each water wave includes:

[0061] The positions of the reflected water waves and the unreflected water waves are marked as z∈[1,m];

[0062]

[0063] In the formula, is the position of the i-th water wave after interference, is the performance index corresponding to the z-th water wave, is the performance index corresponding to the z′th water wave, z′∈[1,m].

[0064] Furthermore, the method for adjusting parameters of the actuator corresponding to the outlier data includes:

[0065] Different digital labels are set for all actuators and marked as execution labels; the actuator corresponding to the outlier data is marked as an outlier, and the execution label corresponding to the outlier is obtained and marked as an outlier label; each outlier label and the corresponding actuator parameter are taken as a set of execution sets, and the execution sets and outlier labels correspond one to one; each outlier data and the corresponding data in the obtained data set are taken as a set of outlier sets, and the outlier sets and outlier data correspond one to one; each execution set is merged with the corresponding outlier set to obtain the corresponding merged set; each merged set is input into the trained parameter adjustment model respectively to predict the corresponding parameter adjustment amount; according to the predicted parameter adjustment amount, the actuator parameters corresponding to the corresponding outlier are adjusted; the training process of the parameter adjustment model is consistent with the training process of the efficiency prediction model, and both are deep neural network models.

[0066] The technical effects and advantages of the PLC-based regulation and management system of the present invention are as follows:

[0067] By collecting industrial status data in real time and using anomaly detection and outlier detection methods, abnormal situations in the industrial production process can be discovered in a timely manner, laying the foundation for subsequent adjustment and optimization; compared with the traditional PID control method, the dynamic optimization algorithm based on deep learning can adaptively optimize the actuator parameters and improve the accuracy and robustness of parameter adjustment; at the same time, by predicting production efficiency and production quality, it can realize intelligent evaluation of the adjustment process to ensure production safety and product quality; it can effectively improve the monitoring and adjustment capabilities in the industrial production process to ensure that industrial equipment is in the best operating state; and enhance the response capability to potential faults and abnormal situations, thereby reducing safety hazards; it promotes the development of industrial production processes in a more automated and intelligent direction, bringing significant results in improving production efficiency, reducing operating costs, and ensuring safe production. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a schematic diagram of a regulation management system based on PLC control according to Example 1 of the present invention;

[0069] Figure 2 This is a flow chart of the outlier data dynamic optimization method according to Example 1 of the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0071] Example 1

[0072] See also Figure 1 As shown, the PLC-controlled regulation and management system described in this embodiment includes a data acquisition module, a data detection module, a state evaluation module, a data optimization module and a control and regulation module; each module is connected by wired and / or wireless means to realize data transmission between modules.

[0073] Data acquisition module, used to collect industrial status data in real time;

[0074] Industrial status data includes equipment operation data, process data and production environment data; equipment operation data such as equipment temperature, equipment pressure, equipment current, etc.; process data such as process temperature, flow, liquid level, etc.; production environment data such as ambient temperature, gas concentration, etc.; It should be understood that the purpose of collecting industrial status data is that the industrial status data can reflect the actual operating status of the industrial production process. In order to ensure the safety and stability of the industrial production process and improve production efficiency and product quality, it is necessary to monitor and control the industrial status data in real time to promptly discover potential equipment failures, process abnormalities and environmental risks, and avoid production interruptions or safety accidents.

[0075] The data detection module is used to detect anomalies in industrial status data and re-collect abnormal data.

[0076] The steps for anomaly detection of industrial status data include:

[0077] Step S101: setting a digital label for each data in the industrial status data in turn and marking it as a data label, the range of the data label is [1, θ], θ is the number of data in the industrial status data;

[0078] Step S102: Select data with data label υ for anomaly detection, υ∈[1,θ];

[0079] Step S103: let υ=υ+1;

[0080] Step S104: looping steps S102 to S103 until all data in the industrial status data have completed abnormality detection, and the loop ends.

[0081] In the above step S102, the method for abnormality detection includes:

[0082] The data to be detected for anomaly are marked as elements; a historical elements are obtained, where a is an integer greater than 1, and the historical elements are obtained by adjusting the database integrated in the management system; a preset analysis quantity b is preset by technical personnel in the field according to actual conditions; the elements collected in real time are marked as real-time elements, and the real-time elements and historical elements are used as analysis data; the corresponding neighbor elements are subtracted from each element in the analysis data to obtain the element difference corresponding to each element; wherein the neighbor elements corresponding to the element are the remaining a elements in the analysis data; all element differences corresponding to each element are regarded as an element set, and the element set corresponds to the element one by one; the element differences in each element set are sorted from small to large to generate an element sorting table corresponding to each element set; the element difference ranked in the bth position in each element sorting table is used as the reference distance of the element corresponding to the corresponding element set, and the neighbor elements corresponding to the first b element differences in each element sorting table are used as the standard elements of the element corresponding to the corresponding element set.

[0083] According to the reference distance and element difference of each element, the adjacent distance between each element and the corresponding standard element is determined; the expression of the adjacent distance is: xl(p,q)=max(jl(p),yc(p,q)); where xl(p,q) is the adjacent distance between the pth element in the analysis data and the corresponding qth standard element, max is the maximum value function, jl(p) is the reference distance of the pth element in the analysis data, yc(p,q) is the element difference between the pth element in the analysis data and the corresponding qth standard element, p∈[1,a+1], q∈[1,b], a>b;

[0084] According to the adjacent distance between each element and the corresponding standard element, the local density of each element is calculated. The expression of local density is: Where md(p) is the local density of the pth element in the analysis data;

[0085] According to the adjacent distance between each element and the corresponding standard element, the abnormal coefficient corresponding to the real-time element is calculated; the expression of the abnormal coefficient is: Where yx is the anomaly coefficient corresponding to the real-time element, md′ is the local density corresponding to the real-time element, and md(q) is the local density of the qth canonical element corresponding to the real-time element.

[0086] A preset quantity range [b, b+c] is preset by technical personnel in this field according to actual conditions; the analysis quantity is set to each value in the quantity range in turn, and the abnormal coefficient corresponding to each value in the quantity range is calculated; all abnormal coefficients are added in turn, and then divided by c+1 to obtain the abnormal mean; a preset judgment coefficient d is set, 0<d<1, and the judgment coefficient d is preset by technical personnel in this field according to actual conditions; if |yz-1|≥d, the data corresponding to the real-time element is marked as abnormal data, and yz is the abnormal mean; if |yz-1|<d, the data corresponding to the real-time element is not marked.

[0087] It should be noted that when step S102 is performed for the first time, υ=1.

[0088] The status evaluation module is used to evaluate the industrial status data and determine whether to generate an adjustment instruction.

[0089] The steps of determining whether to generate an adjustment instruction include:

[0090] Step S201: Select data with data label υ for outlier detection;

[0091] Step S202: let υ=υ+1;

[0092] Step S203: looping steps S201 to S202 until all data in the industrial status data have completed outlier detection, the loop ends, and the process proceeds to step S204;

[0093] Step S204: determine whether there is data marked as outlier data in the industrial status data, if yes, generate an adjustment instruction, if no, do not generate an adjustment instruction.

[0094] In the above step S201, the outlier detection method includes:

[0095] The data for outlier detection is marked as nodes, and the nodes collected in real time are marked as real-time nodes; g historical nodes are obtained, where the historical nodes are normal nodes collected at historical moments, and the normal nodes are nodes within the normal range. g is an integer greater than 1, and the historical nodes are obtained by adjusting the database integrated in the management system; the g historical nodes are added in sequence, and then divided by g to obtain the mean vector; according to the mean vector, the covariance matrix corresponding to the historical node is calculated; the expression of the covariance matrix is: Where Σ is the covariance matrix, C h is the hth node, (C h -μ) T For (C h -μ), h∈[1,g]; according to the covariance matrix, calculate the covariance distance corresponding to the real-time node; the expression of the covariance distance is: Where D is the covariance distance, It is a real-time node;

[0096] A significance level is set, which is pre-set by a technician in this field according to actual conditions, usually 0.01 (indicating a confidence level of 1%) or 0.05 (indicating a confidence level of 5%); the data dimension corresponding to the node is obtained, and the data dimension is the number of types of data in the node; according to the significance level and the data dimension, a distance threshold is obtained by consulting a chi-square distribution table; the covariance distance is squared to obtain the square distance, and the square distance is compared with the distance threshold; if the square distance is greater than the distance threshold, the data corresponding to the corresponding node is marked as outlier data; if the square distance is less than or equal to the distance threshold, the data corresponding to the corresponding node is not marked.

[0097] The data optimization module is used to mark the industrial status data corresponding to the adjustment instruction as outlier data and dynamically optimize the outlier data.

[0098] like Figure 2 As shown in FIG, the steps of dynamically optimizing outlier data include:

[0099] Step S301: Obtain the normal range corresponding to each outlier data;

[0100] Step S302: randomly selecting a value from the normal range corresponding to each outlier data to construct a set of data sets, constructing a total of M sets of data sets, and the M sets of data sets are all different; setting a different digital label for each set of data sets and marking it as a set label;

[0101] Step S303: construct a wave group S, which includes m water waves, the position of each water wave corresponds to the set label one by one, and set the number of iterations t to 0;

[0102] Step S304: defining a performance indicator function and an iteration threshold T;

[0103] Step S305: Divide the wave group S into u sub-wave groups;

[0104] Step S306: propagate each water wave and update the position of each water wave;

[0105] Step S307: Calculate the reflection probability of each water wave and update the position of the reflecting water wave;

[0106] Step S308: interfering each water wave and updating the position of each water wave;

[0107] Step S309: compare the number of iterations t with the iteration threshold T; if t≥T, proceed to step S310; if t<T, set t=t+1 and return to step S306;

[0108] Step S310: Calculate the performance index of each water wave, obtain the water wave with the largest performance index, and mark it as the best water wave; dynamically optimize the outlier data according to the data set corresponding to the set label corresponding to the best water wave.

[0109] In the above step S301, the method for obtaining the normal range corresponding to each outlier data includes:

[0110] The historical nodes corresponding to each outlier data are taken as a set of historical sets, and the historical sets correspond to the outlier data one by one; the historical node with the largest value in each historical set is taken as the maximum value of the corresponding outlier data; the historical node with the smallest value in each historical set is taken as the minimum value of the corresponding outlier data; based on the maximum and minimum values ​​of each outlier data, the normal range corresponding to each outlier data is constructed.

[0111] In the above step S303, the position of each water wave in the wave group S is defined in a one-dimensional search space, and the range of the one-dimensional search space is the set label corresponding to the M group of data sets; the expression of the position of each water wave in the wave group S is: In the formula, is the position of the i-th water wave, P i is the random coefficient of the ith water wave, P i ∈[0,1],i∈[1,m].

[0112] In the above step S304, the expression of the performance index function is: f=ω1×sx+ω2×sz; wherein f is the performance index, sx is the production efficiency, sz is the production quality, ω1 and ω2 are preset weight coefficients; wherein the specific values ​​of the weight coefficients can be set according to actual conditions, and the weight coefficients reflect the degree of influence of production efficiency and production quality on the performance index. Technical personnel in this field can preset corresponding weight coefficients according to the actual degree of influence of production efficiency and production quality on the performance index, so as to accurately evaluate the comprehensive performance of industrial production; since the performance index is only used to evaluate the comprehensive performance of industrial production, it does not have a dimension, that is, the performance index calculation process is a dimensionless calculation.

[0113] The method for obtaining production efficiency and production quality is as follows: obtain the data set corresponding to the set label corresponding to the water wave position, and mark it as a replacement set; replace the outlier data in the industrial status data with the corresponding values ​​in the replacement set, and mark the replaced industrial status data as predicted data; input the predicted data into the trained efficiency prediction model to predict the corresponding production efficiency; input the predicted data into the trained quality prediction model to predict the corresponding production quality; the training process of the efficiency prediction model and the quality prediction model are the same, and both are deep neural network models.

[0114] The training process of the efficiency prediction model includes:

[0115] Pre-collect j groups of prediction data, set corresponding production efficiencies for the j groups of prediction data, where j is an integer greater than 1, and convert the prediction data and the corresponding production efficiencies into a corresponding set of feature vectors; the production efficiencies corresponding to the prediction data are obtained by technicians skilled in the art in the historical process, by collecting j groups of prediction data, carrying out industrial production under the conditions of each group of prediction data, sequentially evaluating the production efficiency in each industrial production process, and using the production efficiency corresponding to the prediction data as the production efficiency, and sequentially setting the corresponding production efficiencies for the j groups of prediction data;

[0116] Each set of feature vectors is used as the input of the efficiency prediction model. The efficiency prediction model uses a set of predicted production efficiency corresponding to each set of predicted data as output, and uses the actual production efficiency corresponding to each set of predicted data as the prediction target. The actual production efficiency is the pre-set production efficiency corresponding to the predicted data. Minimizing the sum of the prediction errors of all predicted data is used as the training target. The calculation formula of the prediction error is η w =(β w -ε w ) 2 , where η w is the prediction error, w is the group number of the feature vector corresponding to the predicted data, β w is the predicted production efficiency corresponding to the wth group of predicted data, ε w is the actual production efficiency corresponding to the wth group of prediction data; the efficiency prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped.

[0117] The above efficiency prediction model is specifically a deep neural network model; it includes an input layer, a hidden layer and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the neurons in the next layer, and the connection contains weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer, and the activation function introduces nonlinearity, allowing the network to learn more complex patterns and features.

[0118] The iteration threshold T is preset by those skilled in the art according to actual conditions.

[0119] In the above step S305, the method of dividing the wave group S into u sub-wave groups includes:

[0120] Calculate the performance index corresponding to each water wave in the wave group S, and sort them from large to small; set an increasing serial number for each water wave in the positive order of sorting, and the serial number range is [1, m]; according to u sub-wave groups, perform a modulo operation on the serial number of each water wave to obtain the corresponding sub-sequence number; the expression of the sub-sequence number is: zx=xl%u; where zx is the sub-sequence number, xl is the sequence number, and % is the modulo function; if the sub-sequence number is not 0, the corresponding water wave is assigned to the zxth sub-wave group; if the sub-sequence number is 0, the corresponding water wave is assigned to the uth sub-wave group; each sub-wave group includes k water waves, and m=uk.

[0121] In the above step S306, the method for updating the position of each water wave includes:

[0122]

[0123] In the formula, is the position of the i-th water wave after propagation, is the position of the i-th water wave before propagation, is the position of the local optimal water wave, X b is the position of the global optimal water wave, is the position of the wave with the largest performance index in the sub-wave group corresponding to the i-th wave, and E1 and E2 are both random numbers between [0,1]. The local optimal wave is the wave with the largest performance index in this iteration, and the global optimal wave is the wave with the largest performance index in all iterations.

[0124] In the above step S307, the expression of reflection probability is: In the formula, is the reflection probability of the i-th water wave, e is a natural constant, f i t is the performance index corresponding to the i-th water wave, f b is the performance index corresponding to the global optimal water wave.

[0125] For each water wave, a random number L between [0,1] is generated i ,like Then the corresponding water wave is marked as reflected water wave. The corresponding water waves are not marked, and the water waves marked as reflected water waves are reflected.

[0126] Methods for updating the position of the reflecting water wave include:

[0127]

[0128] In the formula, is the position of the nth water wave after reflection, is the position of the nth water wave before reflection, y is the reflection factor, l is the propagation factor, is the performance index corresponding to the nth water wave being reflected, n∈[1,N], N is the number of reflected water waves, and r is the gradient factor; the propagation factor and the gradient factor are both preset by technical personnel in this field according to actual conditions.

[0129] In the above step S308, the method for updating the position of each water wave includes:

[0130] The positions of the reflected water waves and the unreflected water waves are marked as z∈[1,m];

[0131]

[0132] In the formula, is the position of the i-th water wave after interference, is the performance index corresponding to the z-th water wave, is the performance index corresponding to the z′th water wave, z′∈[1,m].

[0133] The control and adjustment module is used to adjust the parameters of the actuator corresponding to the outlier data.

[0134] Actuators include valves (used to control flow, liquid level, etc.), heaters (used to control process temperature, ambient temperature, etc.), motors (used to control equipment current, equipment power, etc.), etc.; actuator parameters include the corresponding opening of the valve, the corresponding power of the heater, the corresponding speed of the motor, etc.

[0135] The method for adjusting parameters of the actuator corresponding to the outlier data includes:

[0136] Different digital labels are set for all actuators and marked as execution labels; the actuator corresponding to the outlier data is marked as an outlier, and the execution label corresponding to the outlier is obtained and marked as an outlier label; each outlier label and the corresponding actuator parameter are taken as a set of execution sets, and the execution sets and outlier labels correspond one to one; each outlier data and the corresponding data in the obtained data set are taken as a set of outlier sets, and the outlier sets and outlier data correspond one to one; each execution set is merged with the corresponding outlier set to obtain the corresponding merged set; each merged set is input into the trained parameter adjustment model respectively to predict the corresponding parameter adjustment amount; according to the predicted parameter adjustment amount, the actuator parameters corresponding to the corresponding outlier are adjusted; the training process of the parameter adjustment model is consistent with the training process of the efficiency prediction model, and both are deep neural network models.

[0137] This embodiment collects industrial status data in real time and adopts anomaly detection and outlier detection methods to timely discover abnormal situations in the industrial production process, laying the foundation for subsequent adjustment and optimization; compared with the traditional PID control method, the dynamic optimization algorithm based on deep learning is adopted, which can adaptively optimize the actuator parameters and improve the accuracy and robustness of parameter adjustment; at the same time, by predicting production efficiency and production quality, it realizes intelligent evaluation of the adjustment process to ensure production safety and product quality; it can effectively improve the monitoring and adjustment capabilities in the industrial production process to ensure that industrial equipment is in the best operating state; and enhance the response capability to potential faults and abnormal situations, thereby reducing safety hazards; it promotes the development of industrial production processes in a more automated and intelligent direction, and brings significant results in improving production efficiency, reducing operating costs, and ensuring safe production.

[0138] Example 2

[0139] The present application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by the one or more processors, the regulation management system based on PLC control as described above may be executed.

[0140] The method or system according to the implementation mode of the present application can also be implemented with the help of the architecture of the electronic device shown in the present application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, can store a PLC-controlled regulation management system provided in the present application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in the present application is only exemplary. When implementing different devices, one or more components in the electronic device shown in the present application may be omitted according to actual needs.

[0141] Example 3

[0142] One embodiment of the present application discloses a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by a processor, a PLC-controlled regulation management system according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0143] In addition, according to the embodiments of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided by the present application, for example: a regulation management system based on PLC control. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0144] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0145] 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.

Claims

1. A regulation and management system based on PLC control, characterized in that: include: Data acquisition module, used to collect industrial status data in real time; Data detection module, used to detect anomalies in industrial status data and re-collect abnormal data; A status evaluation module is used to evaluate industrial status data and determine whether to generate adjustment instructions; A data optimization module is used to mark the industrial status data corresponding to the adjustment instruction as outlier data and dynamically optimize the outlier data; The control and adjustment module is used to adjust the parameters of the actuator corresponding to the outlier data.

2. A PLC-controlled regulation and management system according to claim 1, characterized in that: The industrial status data includes equipment operation data, process data and production environment data; The step of performing anomaly detection on industrial status data comprises: Step S101: setting a digital label for each data in the industrial status data in turn and marking it as a data label, the range of the data label is [1, θ], θ is the number of data in the industrial status data; Step S102: Select data with data label υ for anomaly detection, υ∈[1,θ]; Step S103: let υ=υ+1; Step S104: looping steps S102 to S103 until all data in the industrial status data have completed abnormality detection, and the loop ends; In step S102, the method for abnormality detection includes: Mark the data for anomaly detection as elements; obtain a historical elements, where the historical elements are elements collected at historical moments, and a is an integer greater than 1; preset the analysis quantity b, mark the elements collected in real time as real-time elements, and use the real-time elements and historical elements as analysis data; subtract the corresponding neighbor elements from each element in the analysis data to obtain the element difference corresponding to each element; wherein the neighbor elements corresponding to the element are the remaining a elements in the analysis data; determine the reference distance and standard element corresponding to each element based on the analysis quantity and element difference; determine the adjacent distance between each element and each corresponding standard element based on the reference distance and element difference of each element; calculate the local density of each element based on the adjacent distance between each element and each corresponding standard element; calculate the anomaly coefficient corresponding to the real-time element based on the adjacent distance between each element and each corresponding standard element; Preset the quantity range [b, b+c], set the analysis quantity to each value in the quantity range in turn, and calculate the abnormal coefficient corresponding to each value in the quantity range; add all the abnormal coefficients in turn, and then divide them by c+1 to obtain the abnormal mean; preset the judgment coefficient d, 0<d<1; if |yz-1|≥d, the data corresponding to the real-time element is marked as abnormal data, yz is the abnormal mean; if |yz-1|<d, the data corresponding to the real-time element is not marked.

3. A PLC-controlled regulation and management system according to claim 2, characterized in that: The method for determining the reference distance and the standard element corresponding to each element comprises: All element differences corresponding to each element are taken as an element set, and the element set corresponds to the element one by one; the element differences in each element set are sorted from small to large to generate an element sorting table corresponding to each element set; the element difference ranked in the bth position in each element sorting table is taken as the reference distance of the element corresponding to the corresponding element set, and the neighbor element corresponding to the first b element differences in each element sorting table is taken as the standard element of the element corresponding to the corresponding element set; The expression of the adjacent distance is: xl(p,q)=max(jl(p),yc(p,q)); where xl(p,q) is the adjacent distance between the p-th element in the analysis data and the corresponding q-th standard element, max is the maximum function, jl(p) is the reference distance of the p-th element in the analysis data, yc(p,q) is the element difference between the p-th element in the analysis data and the corresponding q-th standard element, p∈[1,a+1], q∈[1,b], a>b; The expression of local density is: Where md(p) is the local density of the pth element in the analysis data; The expression of the anomaly coefficient is: Where yx is the anomaly coefficient corresponding to the real-time element, md′ is the local density corresponding to the real-time element, and md(q) is the local density of the qth canonical element corresponding to the real-time element.

4. A PLC-controlled regulation and management system according to claim 3, characterized in that: The step of determining whether to generate an adjustment instruction comprises: Step S201: Select data with data label υ for outlier detection; Step S202: let υ=υ+1; Step S203: looping steps S201 to S202 until all data in the industrial status data have completed outlier detection, the loop ends, and the process proceeds to step S204; Step S204: determining whether there is data in the industrial status data that is marked as outlier data, if so, generating an adjustment instruction, if not, not generating an adjustment instruction; In step S201, the outlier detection method includes: Mark the data for outlier detection as nodes, and mark the nodes collected in real time as real-time nodes; obtain g historical nodes, where the historical nodes are normal nodes collected at historical moments, the normal nodes are nodes within the normal range, and g is an integer greater than 1; add the g historical nodes in sequence, and then divide by g to obtain the mean vector; calculate the covariance matrix corresponding to the historical nodes based on the mean vector; calculate the covariance distance corresponding to the real-time nodes based on the covariance matrix; Set the significance level; obtain the data dimension corresponding to the node, where the data dimension is the number of data types in the node; obtain the distance threshold based on the significance level and data dimension; square the covariance distance to obtain the square distance, and compare the square distance with the distance threshold; if the square distance is greater than the distance threshold, mark the data corresponding to the corresponding node as outlier data; if the square distance is less than or equal to the distance threshold, do not mark the data corresponding to the corresponding node.

5. A PLC-controlled regulation and management system according to claim 4, characterized in that: The expression of the covariance matrix is: Where Σ is the covariance matrix, C h is the hth node, (C h -μ) T For (C h -μ), h∈[1,g]; The expression of covariance distance is: Where D is the covariance distance and C is the real-time node.

6. A PLC-controlled regulation and management system according to claim 5, characterized in that: The step of dynamically optimizing outlier data comprises: Step S301: Obtain the normal range corresponding to each outlier data; Step S302: randomly selecting a value from the normal range corresponding to each outlier data to construct a set of data sets, constructing a total of M sets of data sets, and the M sets of data sets are all different; setting a different digital label for each set of data sets and marking it as a set label; Step S303: construct a wave group S, which includes m water waves, the position of each water wave corresponds to the set label one by one, and set the number of iterations t to 0; Step S304: defining a performance indicator function and an iteration threshold T; Step S305: Divide the wave group S into u sub-wave groups; Step S306: propagate each water wave and update the position of each water wave; Step S307: Calculate the reflection probability of each water wave and update the position of the reflecting water wave; Step S308: interfering each water wave and updating the position of each water wave; Step S309: compare the number of iterations t with the iteration threshold T; if t≥T, proceed to step S310; if t<T, set t=t+1 and return to step S306; Step S310: Calculate the performance index of each water wave, obtain the water wave with the largest performance index, and mark it as the best water wave; dynamically optimize the outlier data according to the data set corresponding to the set label corresponding to the best water wave.

7. A PLC-controlled regulation and management system according to claim 6, characterized in that: In step S301, the method for obtaining the normal range corresponding to each outlier data includes: The historical nodes corresponding to each outlier data are taken as a set of historical sets, and the historical sets correspond to the outlier data one by one; the historical node with the largest value in each historical set is taken as the maximum value of the corresponding outlier data; the historical node with the smallest value in each historical set is taken as the minimum value of the corresponding outlier data; based on the maximum and minimum values ​​of each outlier data, the normal range corresponding to each outlier data is constructed; In step S303, the position of each water wave in the wave group S is defined in a one-dimensional search space, and the range of the one-dimensional search space is the set label corresponding to the M data sets; the expression of the position of each water wave in the wave group S is: In the formula, is the position of the i-th water wave, P i is the random coefficient of the ith water wave, P i ∈[0, 1], i∈[1,m]; In the step S304, the expression of the performance index function is: f=ω1×sx+ω2×sz; wherein f is the performance index, sx is the production efficiency, sz is the production quality, ω1 and ω2 are preset weight coefficients; the method for obtaining the production efficiency and production quality is: obtaining the data set corresponding to the set label corresponding to the water wave position, and marking it as a replacement set; replacing all outlier data in the industrial status data with corresponding values ​​in the replacement set, and marking the replaced industrial status data as predicted data; inputting the predicted data into the trained efficiency prediction model to predict the corresponding production efficiency; inputting the predicted data into the trained quality prediction model to predict the corresponding production quality; the training processes of the efficiency prediction model and the quality prediction model are consistent, and both are deep neural network models.

8. A PLC-controlled regulation and management system according to claim 7, characterized in that: In step S305, the method of dividing the wave group S into u sub-wave groups includes: Calculate the performance index corresponding to each water wave in the wave group S, and sort them from large to small; set an increasing serial number for each water wave in the positive order of sorting, and the serial number range is [1, m]; according to u sub-wave groups, perform a modulo operation on the serial number of each water wave to obtain the corresponding sub-sequence number; the expression of the sub-sequence number is: zx = xl%u; where zx is the sub-sequence number, xl is the sequence number, and % is the modulo function; if the sub-sequence number is not 0, the corresponding water wave is assigned to the zxth sub-wave group; if the sub-sequence number is 0, the corresponding water wave is assigned to the uth sub-wave group; each sub-wave group includes k water waves, and m = uk; In step S306, the method for updating the position of each water wave includes: In the formula, is the position of the i-th water wave after propagation, is the position of the i-th water wave before propagation, is the position of the local optimal water wave, X b is the position of the global optimal water wave, is the position of the wave with the largest performance index in the sub-wave group corresponding to the i-th wave, and E1 and E2 are both random numbers between [0,1]. The local optimal wave is the wave with the largest performance index in this iteration, and the global optimal wave is the wave with the largest performance index in all iterations.

9. A regulation management system based on PLC control according to claim 8, characterized in that: In step S307, the expression of reflection probability is: In the formula, is the reflection probability of the i-th water wave, e is a natural constant, f i t is the performance index corresponding to the i-th water wave, f b is the performance index corresponding to the global optimal water wave; For each water wave, a random number L between [0,1] is generated i ,like Then the corresponding water wave is marked as reflected water wave. The corresponding water waves are not marked, and the water waves marked as reflected water waves are reflected; Methods for updating the position of the reflecting water wave include: In the formula, is the position of the nth water wave after reflection, is the position of the nth water wave before reflection, y is the reflection factor, l is the propagation factor, is the performance index corresponding to the nth water wave being reflected, n∈[1,N], N is the number of reflected water waves, and r is the gradient factor; In step S308, the method for updating the position of each water wave includes: The positions of the reflected water waves and the unreflected water waves are marked as z∈[1,m]; In the formula, is the position of the i-th water wave after interference, is the performance index corresponding to the z-th water wave, is the performance index corresponding to the z′th water wave, z′∈[1,m].

10. A regulation management system based on PLC control according to claim 9, characterized in that: The method for adjusting parameters of the actuator corresponding to the outlier data includes: Different digital labels are set for all actuators and marked as execution labels; the actuator corresponding to the outlier data is marked as an outlier, and the execution label corresponding to the outlier is obtained and marked as an outlier label; each outlier label and the corresponding actuator parameter are taken as a set of execution sets, and the execution sets and outlier labels correspond one to one; each outlier data and the corresponding data in the obtained data set are taken as a set of outlier sets, and the outlier sets and outlier data correspond one to one; each execution set is merged with the corresponding outlier set to obtain the corresponding merged set; each merged set is input into the trained parameter adjustment model respectively to predict the corresponding parameter adjustment amount; according to the predicted parameter adjustment amount, the actuator parameters corresponding to the corresponding outlier are adjusted; the training process of the parameter adjustment model is consistent with the training process of the efficiency prediction model, and both are deep neural network models.

Images