Precise control system and method for intelligent heating digestion furnace

Abstract

The invention discloses a precise control system and method for an intelligent heating digestion furnace, relates to the field of intelligent control, and adopts an artificial intelligence technology based on deep learning to carry out real-time monitoring and data analysis on in-furnace temperature and heating power of the digestion furnace. According to the method, the time sequence change trend of the offset of the in-furnace temperature value relative to the set temperature value and the time sequence change trend of the heating power are extracted, and then the dynamic response mode of the in-furnace temperature relative to the heating power is excavated by performing fine-grained time sequence interaction analysis on the in-furnace temperature value and the time sequence change trend, so that intelligent regulation and control on the heating power are realized. Through the mode, the temperature control stability of the heating digestion furnace can be effectively improved, the energy consumption is reduced, and the efficiency and the reliability of the digestion process are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to a precise control system and method for an intelligent heating and digestion furnace. Background Art

[0002] In many fields such as chemical analysis, environmental monitoring, materials science and biomedicine, heating digestion furnaces are widely used as key experimental equipment in sample pretreatment processes, such as organic matter digestion, inorganic matter dissolution, sample ashing, etc. Since any slight temperature fluctuation may have a significant impact on the experimental results, and even lead to experimental failure or data distortion, these processes have extremely high requirements for the accuracy and stability of temperature control.

[0003] The temperature control of traditional heating digestion furnaces mostly relies on a simple PID (proportional-integral-differential) control algorithm. Although this method can achieve stable temperature control to a certain extent, the setting of PID control algorithm parameters (proportional, integral, differential) often relies on experience and trial and error methods, and the heating digestion furnace usually has large hysteresis and nonlinear time-varying characteristics, which makes it difficult to accurately set the parameters of the PID controller, and is prone to problems such as temperature overshoot, oscillation or slow adjustment, thus failing to meet the high-precision and high-stability requirements of modern experiments for temperature control.

[0004] Therefore, an optimized intelligent heating digestion furnace precision control system and method are expected. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed.

[0006] According to one aspect of the present application, a precise control method of an intelligent heating digestion furnace is provided, which includes: Obtain the time queue of the real-time temperature in the digestion furnace collected by the temperature sensor, and set the temperature value; Calculate the difference between each real-time temperature in the digestion furnace in the time queue of the real-time temperature in the digestion furnace and the set temperature value to obtain a time queue of the real-time temperature target offset; A time queue for obtaining the real-time power of the heating element collected by the power sensor; Respectively extracting the time series variation characteristics of the time series of the real-time temperature target offset and the time series of the real-time power to obtain a real-time temperature target offset time series correlation feature vector and a heating element power time series correlation feature vector; Performing local fine-grained interactive response analysis on the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; Based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, a power regulation instruction is determined.

[0007] In the precise control method of the above-mentioned intelligent heating digestion furnace, the timing change characteristics of the time queue of the real-time temperature target offset and the time queue of the real-time power are extracted respectively to obtain the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector, including: the time queue of the real-time temperature target offset and the time queue of the real-time power are respectively input into a sequence encoder based on a Bi-LSTM model to obtain the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector.

[0008] In the precise control method of the above-mentioned intelligent heating digestion furnace, a local fine-grained interactive response analysis is performed on the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, including: inputting the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector into a local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0009] In the precise control method of the above-mentioned intelligent heating digestion furnace, the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector are input into a local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, including: feature decoupling the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector respectively to obtain a sequence of real-time temperature target offset timing feature local representation vectors and a sequence of heating element power timing feature local representation vectors; calculating the real-time temperature target offset timing feature local representation vector sequence and the heating element power timing feature local representation vector sequence corresponding to each group of the real-time temperature target offset timing feature local representation vector sequence. The local semantic association interaction vector between the feature local representation vector and the heating element power timing feature local representation vector is used to obtain a sequence of temperature offset-heating power local semantic association interaction vectors; the fine-grained attention weights of each temperature offset-heating power local semantic association interaction vector in the sequence of temperature offset-heating power local semantic association interaction vectors are calculated to obtain a sequence of fine-grained attention weights; the sequence of fine-grained attention weights is input into a random deactivation module to obtain a sequence of fine-grained attention weights after deactivation processing; the sequence of fine-grained attention weights after deactivation processing is used as the weight distribution, and the weighted sum of the sequence of temperature offset-heating power local semantic association interaction vectors is calculated to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0010] In the precise control method of the above-mentioned intelligent heating digestion furnace, the local semantic association interaction vector between each corresponding group of the real-time temperature target offset timing feature local representation vector and the heating element power timing feature local representation vector in the sequence of the real-time temperature target offset timing feature local representation vector and the sequence of the heating element power timing feature local representation vector is calculated to obtain a sequence of temperature offset-heating power local semantic association interaction vectors, including: calculating the position point addition, position point subtraction and position point multiplication between the real-time temperature target offset timing feature local representation vector and the corresponding heating element power timing feature local representation vector to obtain a first fusion result, a second fusion result and a third fusion result; after cascading the first fusion result, the second fusion result and the third fusion result into a temperature offset-heating power multi-scale fusion feature vector, one-dimensional convolution processing and maximum pooling processing are performed on them to obtain the temperature offset-heating power local semantic association interaction vector.

[0011] In the precise control method of the above-mentioned intelligent heating digestion furnace, the fine-grained attention weights of each temperature offset-heating power local semantic association interaction vector in the sequence of the temperature offset-heating power local semantic association interaction vectors are calculated to obtain a sequence of fine-grained attention weights, including: multiplying the temperature offset-heating power local semantic association interaction vector by a weight coefficient matrix, adding it to the bias vector and then passing it through a tanh function to obtain a temperature offset-heating power local semantic association interaction encoding feature vector; using an attention score reference vector multiplied by the temperature offset-heating power local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

[0012] In the precise control method of the above-mentioned intelligent heating digestion furnace, the power adjustment instruction is determined based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, including: inputting the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector into a classifier-based power regulator to obtain the power adjustment instruction, and the power adjustment instruction is used to indicate that the heating power at the next time point should increase, decrease or remain unchanged.

[0013] In the precise control method of the above-mentioned intelligent heating digestion furnace, the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector is input into a power regulator based on a classifier to obtain the power adjustment instruction, and the power adjustment instruction is used to indicate whether the heating power at the next time point should increase, decrease or remain unchanged, including: using multiple fully connected layers of the classifier to fully connect encode the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector to obtain an encoded classification feature vector; passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain the power adjustment instruction.

[0014] According to another aspect of the present application, a precise control system for an intelligent heating and digestion furnace is provided, which includes: The temperature data acquisition module is used to obtain the time queue of the real-time temperature in the digestion furnace collected by the temperature sensor, and set the temperature value; A real-time temperature target offset calculation module, used for calculating the difference between each real-time temperature in the digestion furnace and the set temperature value in the time queue of the real-time temperature in the digestion furnace to obtain a time queue of the real-time temperature target offset; A power data acquisition module, used to acquire a time queue of the real-time power of the heating element collected by the power sensor; A time series variation feature extraction module, used to extract the time series variation features of the time series of the real-time temperature target offset and the time series of the real-time power respectively to obtain a real-time temperature target offset time series correlation feature vector and a heating element power time series correlation feature vector; A local fine-grained interactive response analysis module, used for performing local fine-grained interactive response analysis on the real-time temperature target offset timing correlation feature vector and the heating element power timing correlation feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; The adjustment instruction determination module is used to determine the power adjustment instruction based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0015] Compared with the prior art, the precise control system and method of an intelligent heating digestion furnace provided in this application uses artificial intelligence technology based on deep learning to perform real-time monitoring and data analysis on the temperature and heating power in the digestion furnace, extract the time series change trend of the offset of the temperature value in the furnace relative to the set temperature value, and the time series change trend of the heating power, and then through fine-grained time series interaction analysis of the two, dig out the dynamic response mode of the temperature in the furnace relative to the heating power, so as to realize intelligent regulation of the heating power. In this way, the temperature control stability of the heating digestion furnace can be effectively improved, energy consumption can be reduced, and the efficiency and reliability of the digestion process can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0017] Figure 1 It is a flow chart of a precise control method of an intelligent heating digestion furnace according to an embodiment of the present application; Figure 2 A data flow diagram of a precise control method of an intelligent heating digestion furnace according to an embodiment of the present application; Figure 3 It is a block diagram of a precise control system of an intelligent heating digestion furnace according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0019] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0020] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0021] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] The temperature control of traditional heating digestion furnaces mostly relies on a simple PID (proportional-integral-differential) control algorithm. Although this method can achieve stable temperature control to a certain extent, the setting of PID control algorithm parameters (proportional, integral, differential) often relies on experience and trial and error methods, and the heating digestion furnace usually has large hysteresis and nonlinear time-varying characteristics, which makes it difficult to accurately set the parameters of the PID controller, and is prone to problems such as temperature overshoot, oscillation or slow adjustment, thus failing to meet the high-precision and high-stability requirements of modern experiments for temperature control. Therefore, an optimized intelligent heating digestion furnace precision control system and method is expected.

[0024] In the technical solution of the present application, a precise control method of an intelligent heating digestion furnace is proposed. Figure 1 It is a flow chart of a precise control method of an intelligent heating digestion furnace according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the precise control method of the intelligent heating digestion furnace according to the embodiment of the present application. Figure 1 and Figure 2As shown, according to the precise control method of the intelligent heating digestion furnace of the embodiment of the present application, the steps include: S1, obtaining the time queue of the real-time temperature in the digestion furnace collected by the temperature sensor, and the set temperature value; S2, calculating the difference between each real-time temperature in the digestion furnace in the time queue of the real-time temperature in the digestion furnace and the set temperature value to obtain the time queue of the real-time temperature target offset; S3, obtaining the time queue of the real-time power of the heating element collected by the power sensor; S4, respectively extracting the timing change characteristics of the time queue of the real-time temperature target offset and the time queue of the real-time power to obtain the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector; S5, performing local fine-grained interactive response analysis on the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; S6, determining the power adjustment instruction based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0025] In particular, the S1 and S2 obtain the time queue of the real-time temperature in the digestion furnace collected by the temperature sensor, and the set temperature value; and calculate the difference between each real-time temperature in the digestion furnace in the time queue of the real-time temperature in the digestion furnace and the set temperature value to obtain the time queue of the real-time temperature target offset. It should be understood that during the digestion process, due to the influence of various factors such as material properties, ambient temperature changes, and heating element efficiency changes, the temperature in the furnace may deviate from the set value. By monitoring the actual temperature in the furnace in real time and comparing it with the set target temperature, the gap between the current state and the target state can be dynamically perceived, and the heating power can be adjusted accordingly to keep the temperature in the furnace stable. Based on this, in order to quantitatively represent the deviation between the actual temperature and the target temperature, the present application further calculates the difference between each real-time temperature in the digestion furnace in the time queue of the real-time temperature in the digestion furnace and the set temperature value to reveal the real-time offset of the temperature in the furnace, thereby obtaining the time queue of the real-time temperature target offset as the basis for regulating the temperature in the furnace.

[0026] In particular, the S3 obtains the time queue of the real-time power of the heating element collected by the power sensor. It should be understood that the power output of the heating element of the digestion furnace directly affects the change of the temperature in the furnace. In order to achieve accurate control of the temperature in the furnace, in the technical solution of the present application, the power output of the heating element is further monitored in real time by the power sensor to obtain the time queue of the real-time power.

[0027] In particular, the S4 extracts the time series change characteristics of the time queue of the real-time temperature target offset and the time queue of the real-time power respectively to obtain the time series associated feature vector of the real-time temperature target offset and the time series associated feature vector of the heating element power. In a specific example of the present application, the time queue of the real-time temperature target offset and the time queue of the real-time power are respectively input into a sequence encoder based on a Bi-LSTM model to obtain the time series associated feature vector of the real-time temperature target offset and the time series associated feature vector of the heating element power. Here, in order to extract the dynamic change trend of the temperature offset and the heating power over time, the present application further adopts a Bi-LSTM model as a sequence encoder to perform time series modeling on the time queue of the real-time temperature target offset and the time queue of the real-time power respectively, so as to utilize the bidirectional information processing capability of the Bi-LSTM model, and simultaneously capture the forward and backward time series context information in the time queue data, and respectively mine the time series dynamic change patterns of the temperature offset and the heating power, so as to obtain the time series associated feature vector of the real-time temperature target offset and the time series associated feature vector of the heating element power. It is worth mentioning that Bi-LSTM is a type of recurrent neural network that processes sequence data by considering both past and future contextual information. It consists of two LSTM layers, one processing the forward direction of the sequence and the other processing the reverse direction of the sequence. The Bi-LSTM model connects two LSTM layers, one processing the forward direction of the sequence and the other processing the reverse direction of the sequence. By considering both past and future contextual information at the same time, the Bi-LSTM model is able to better capture long-term dependencies in the sequence.

[0028] Accordingly, in a possible implementation, the time queue of the real-time temperature target offset and the time queue of the real-time power can be respectively input into a sequence encoder based on a Bi-LSTM model through the following steps to obtain the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector, for example: the time queue of the real-time temperature target offset and the time queue of the real-time power are respectively input into the Bi-LSTM sequence encoder; the Bi-LSTM encoder processes each time step in the sequence and outputs a hidden state; the hidden state of each time step is connected to form an encoded sequence; the last hidden state is selected from the encoded sequence as the timing association feature vector to obtain the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector.

[0029] In particular, the S5 performs a local fine-grained interactive response analysis on the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector. In a specific example of the present application, the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector are input into a local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector. Since the furnace temperature and heating power of the digestion furnace are two mutually influential variables, there is a complex nonlinear correlation relationship between the two. In the technical solution of the present application, in order to fully explore the complex correlation pattern between the temperature offset and the heating power, a local fine-grained sequence interactive response aggregation network based on feature decoupling is further introduced to perform an interactive response analysis on the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector to reveal the influence intensity of the heating power on the temperature offset, and the sensitivity of the furnace temperature to the change of the heating power. Specifically, the network first decouples the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector respectively, decomposing the two into two sets of local representation vector sequences to enhance the sensitivity to local detail features. Then, the local semantic association modeling is performed on the sequence of the two sets of local representation vectors after decoupling to generate a local semantic association interaction vector sequence, and the feature interaction response strength is further quantified to generate an attention weight sequence, so that the model can focus on the most important feature interaction, thereby improving the quality of feature representation and regulation accuracy. At the same time, in order to prevent overfitting, the generated attention weight sequence is further subjected to random inactivation processing to enhance the generalization ability of the model. Finally, the attention weight sequence after random inactivation processing is used as the weight distribution, and the local semantic association interaction vector sequence is weighted summed to synthesize the local interaction information of the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector, and obtain the global temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0030] In an embodiment of the present application, the real-time temperature target offset timing-associated feature vector and the heating element power timing-associated feature vector are input into a local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, including: performing feature decoupling on the real-time temperature target offset timing-associated feature vector and the heating element power timing-associated feature vector respectively to obtain a sequence of real-time temperature target offset timing feature local representation vectors and a sequence of heating element power timing feature local representation vectors; calculating the real-time temperature target offset timing feature local representation vectors and the heating element power timing feature local representation vectors corresponding to each group of the real-time temperature target offset timing feature local representation vectors. The invention relates to a method for realizing a method for realizing a decoupling and reinforcing interactive control inference representation vector of a heating element power by using a local semantic association interaction vector between an indicator vector and a local representation vector of the heating element power timing feature to obtain a sequence of temperature offset-heating power local semantic association interaction vectors; calculating the fine-grained attention weights of each temperature offset-heating power local semantic association interaction vector in the sequence of temperature offset-heating power local semantic association interaction vectors to obtain a sequence of fine-grained attention weights; inputting the sequence of fine-grained attention weights into a random deactivation module to obtain a sequence of fine-grained attention weights after deactivation processing; using the sequence of fine-grained attention weights after deactivation processing as a weight distribution, calculating the weighted sum of the sequence of temperature offset-heating power local semantic association interaction vectors to obtain the temperature offset-heating power decoupling and reinforcing interactive control inference representation vector.

[0031] Among them, the process of calculating the local semantic association interaction vector between each group of corresponding real-time temperature target offset timing feature local representation vectors and heating element power timing feature local representation vectors in the sequence of the real-time temperature target offset timing feature local representation vectors and the sequence of the heating element power timing feature local representation vectors to obtain a sequence of temperature offset-heating power local semantic association interaction vectors includes: calculating the position point addition, position point subtraction and position point multiplication between the real-time temperature target offset timing feature local representation vector and the corresponding heating element power timing feature local representation vector to obtain a first fusion result, a second fusion result and a third fusion result; after cascading the first fusion result, the second fusion result and the third fusion result into a temperature offset-heating power multi-scale fusion feature vector, one-dimensional convolution processing and maximum value pooling processing are performed on them to obtain the temperature offset-heating power local semantic association interaction vector.

[0032] More specifically, the fine-grained attention weights of each temperature offset-heating power local semantic association interaction vector in the sequence of the temperature offset-heating power local semantic association interaction vectors are calculated to obtain a sequence of fine-grained attention weights, including: multiplying the temperature offset-heating power local semantic association interaction vector by a weight coefficient matrix, adding the result to the bias vector and then passing the result through a tanh function to obtain a temperature offset-heating power local semantic association interaction encoding feature vector; using an attention score reference vector multiplied by the temperature offset-heating power local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

[0033] In summary, in the above embodiment, the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector are input into a local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, including: processing the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector with the following interactive response analysis formula to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, wherein the interactive response analysis formula is: ; ; ; ; ; ; ;in, represents the real-time temperature target offset timing associated feature vector, represents the heating element power time series associated feature vector, Indicates decoupling processing, , , and respectively represent the first, second, and third in the sequence of the local representation vector of the real-time temperature target offset timing feature. and A local representation vector of the temporal characteristics of the real-time temperature target offset, The value of is the number of feature vectors in the sequence of the local representation vector of the real-time temperature target offset timing feature, , , and The first, second, and third vectors in the sequence of the local representation vectors of the heating element power timing characteristics are represented respectively. and The local representation vector of the heating element power time series characteristics, Indicates cascade operation, represents the maximum pooling operation, represents a one-dimensional convolution operation, represents dot product, Indicates point addition, Indicates point reduction, The first in the sequence of the temperature offset-heating power local semantic association interaction vector temperature offset-heating power local semantic association interaction vector, is the weight coefficient matrix, is the bias vector, is the hyperbolic tangent function, is the attention score reference vector, Indicates fine-grained attention weights, represents random dropout processing, Indicates The fine-grained attention weights after dropout processing, The temperature offset-heating power decoupling enhanced interactive control reasoning representation vector is represented.

[0034] In particular, the S6 determines the power adjustment instruction based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector. In a specific example of the present application, the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector is input into a power regulator based on a classifier to obtain the power adjustment instruction, and the power adjustment instruction is used to indicate that the heating power at the next time point should increase, decrease, or remain unchanged. Among them, the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector is input into a power regulator based on a classifier to obtain the power adjustment instruction, and the power adjustment instruction is used to indicate that the heating power at the next time point should increase, decrease, or remain unchanged, including: using multiple fully connected layers of the classifier to fully connect encode the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector to obtain an encoded classification feature vector; passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain the power adjustment instruction.

[0035] In a preferred example, inputting the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector into a classifier-based power regulator to obtain a power regulation instruction comprises the steps of: Clustering all eigenvalues ​​of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector based on the distance between eigenvalues, such as the absolute value distance of the difference, and arranging the clustering features into a temperature offset-heating power decoupling enhanced interactive control reasoning clustering vector; Determine a clustering ratio value of the number of eigenvalues ​​of the temperature offset-heating power decoupling enhanced interactive control reasoning cluster vector and the number of eigenvalues ​​of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; Dividing the second norm of the temperature offset-heating power decoupling enhanced interactive control reasoning clustering vector by the second norm of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning conflict representation value; Divide a first power value of a norm of the temperature offset-heating power decoupling enhanced interactive control reasoning cluster vector with the cluster ratio value as an exponent by a second power value of a norm of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector with the cluster ratio value as an exponent to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning adversarial representation value; For each eigenvalue of the temperature offset-heating power decoupling enhanced interactive control reasoning clustering vector, multiply it by the inverse of the difference between the temperature offset-heating power decoupling enhanced interactive control reasoning conflict representation value and the temperature offset-heating power decoupling enhanced interactive control reasoning antagonism representation value to obtain the optimized eigenvalue of the temperature offset-heating power decoupling enhanced interactive control reasoning clustering vector; For each eigenvalue outside the cluster in the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, multiply it by the reciprocal of the sum of the temperature offset-heating power decoupling enhanced interactive control reasoning conflict representation value and the temperature offset-heating power decoupling enhanced interactive control reasoning antagonism representation value to obtain an optimized out-of-class eigenvalue of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; Combining the optimized eigenvalues ​​of the temperature offset-heating power decoupling enhanced interactive control reasoning clustering vector and the optimized out-of-class eigenvalues ​​of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector into an optimized temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; The optimized temperature offset-heating power decoupling enhanced interactive control reasoning representation vector is input into the classifier-based power regulator to obtain a power regulation instruction, which is expressed as: ; ;in, is each eigenvalue of the temperature offset-heating power decoupling enhanced interactive control inference representation vector, is the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, is the number of eigenvalues ​​of the temperature offset-heating power decoupling enhanced interactive control inference representation vector, is the temperature offset-heating power decoupling enhanced interactive control inference clustering vector, is the number of eigenvalues ​​of the temperature offset-heating power decoupling enhanced interactive control inference clustering vector, represents the cluster feature set corresponding to the temperature offset-heating power decoupling enhanced interactive control reasoning cluster vector, and Represents the binary norm and the uninorm of the vector respectively. Power, is each eigenvalue of the optimized temperature offset-heating power decoupling enhanced interactive control inference representation vector.

[0036] That is, considering that the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector represent the short-range-long-range timing association features of the real-time temperature target offset and the real-time power respectively, when further performing local fine-grained sequence interactive response aggregation based on feature decoupling, the local fine-grained sequence interactive responsiveness aggregation alignment conflict will be caused due to the global multi-scale correlation differences of the timing features, thereby causing the loss of aggregation key information and affecting the expression effect of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0037] Therefore, in order to avoid the loss of key suffix semantic information relative to the original feature set as a whole due to aggregation conflict caused by the temperature offset-heating power decoupling and enhanced interactive control reasoning representation vector being based on aggregated features, the clustering ratio of the number of eigenvalues ​​of the temperature offset-heating power decoupling and enhanced interactive control reasoning clustering vector relative to the number of eigenvalues ​​of the temperature offset-heating power decoupling and enhanced interactive control reasoning representation vector is used as a decision function to perform an adversarial decision on the set absolute representation of the norm of the temperature offset-heating power decoupling and enhanced interactive control reasoning clustering vector and the temperature offset-heating power decoupling and enhanced interactive control reasoning representation vector, and respectively compare them with the temperature offset-heating power decoupling and enhanced interactive control reasoning representation vector. The clustering intrinsic conflict representation of the two norm of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector and the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector is positively and negatively interacted to construct a solid alignment guardrail of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector based on the aggregated features and the original feature set as a whole, thereby achieving the intentional mitigation of harmful information loss based on the aggregated risk transferability of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, improving the expression effect of the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, and thus improving the accuracy of the power regulation instruction obtained by the power regulator based on the classifier input by the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0038] In summary, the precise control method of the intelligent heating digestion furnace according to the embodiment of the present application is explained, which uses artificial intelligence technology based on deep learning to perform real-time monitoring and data analysis on the temperature and heating power of the digestion furnace, extracts the time series change trend of the offset of the temperature value in the furnace relative to the set temperature value, and the time series change trend of the heating power, and then through fine-grained time series interaction analysis of the two, digs out the dynamic response mode of the temperature in the furnace relative to the heating power, so as to realize intelligent regulation of the heating power. In this way, the temperature control stability of the heating digestion furnace can be effectively improved, energy consumption can be reduced, and the efficiency and reliability of the digestion process can be improved.

[0039] Furthermore, a precise control system for an intelligent heating and digestion furnace is also provided.

[0040] Figure 3 FIG. 1 is a block diagram of a precise control system of an intelligent heating and digestion furnace according to an embodiment of the present application. Figure 3 As shown, according to the precise control system 300 of the intelligent heating digestion furnace of the embodiment of the present application, it includes: a temperature data acquisition module 310, which is used to obtain the time queue of the real-time temperature in the digestion furnace collected by the temperature sensor, and the set temperature value; a real-time temperature target offset calculation module 320, which is used to calculate the difference between each real-time temperature in the digestion furnace in the time queue of the real-time temperature in the digestion furnace and the set temperature value to obtain the time queue of the real-time temperature target offset; a power data acquisition module 330, which is used to obtain the time queue of the real-time power of the heating element collected by the power sensor; a time series change feature extraction module 340, which is used to respectively extract the real-time temperature target offset value. Take the timing variation characteristics of the time queue of the real-time temperature target offset and the time queue of the real-time power to obtain the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector; a local fine-grained interactive response analysis module 350 is used to perform local fine-grained interactive response analysis on the real-time temperature target offset timing association feature vector and the heating element power timing association feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; an adjustment instruction determination module 360 ​​is used to determine the power adjustment instruction based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

[0041] As described above, the precise control system 300 of the intelligent heating and digestion furnace according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a precise control algorithm of the intelligent heating and digestion furnace. In one possible implementation, the precise control system 300 of the intelligent heating and digestion furnace according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the precise control system 300 of the intelligent heating and digestion furnace can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the precise control system 300 of the intelligent heating and digestion furnace can also be one of the many hardware modules of the wireless terminal.

[0042] Alternatively, in another example, the precise control system 300 of the intelligent heating and digestion furnace and the wireless terminal may be separate devices, and the precise control system 300 of the intelligent heating and digestion furnace may be connected to the wireless terminal via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0043] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A precise control method for an intelligent heating digestion furnace, characterized in that: include: Obtain the time queue of the real-time temperature in the digestion furnace collected by the temperature sensor, and set the temperature value; Calculate the difference between each real-time temperature in the digestion furnace in the time queue of the real-time temperature in the digestion furnace and the set temperature value to obtain a time queue of the real-time temperature target offset; A time queue for obtaining the real-time power of the heating element collected by the power sensor; Respectively extracting the time series variation characteristics of the time series of the real-time temperature target offset and the time series of the real-time power to obtain a real-time temperature target offset time series correlation feature vector and a heating element power time series correlation feature vector; Performing local fine-grained interactive response analysis on the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; Based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, a power regulation instruction is determined.

2. The precise control method of the intelligent heating digestion furnace according to claim 1, characterized in that: Respectively extracting the time series variation characteristics of the time series of the real-time temperature target offset and the time series of the real-time power to obtain a real-time temperature target offset time series correlation feature vector and a heating element power time series correlation feature vector, including: The time queue of the real-time temperature target offset and the time queue of the real-time power are respectively input into a sequence encoder based on a Bi-LSTM model to obtain a time series associated feature vector of the real-time temperature target offset and a time series associated feature vector of the heating element power.

3. The precise control method of the intelligent heating digestion furnace according to claim 2 is characterized in that: Performing local fine-grained interactive response analysis on the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, including: The real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector are input into a local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

4. The precise control method of the intelligent heating digestion furnace according to claim 3 is characterized in that: Inputting the real-time temperature target offset time series associated feature vector and the heating element power time series associated feature vector into a local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector, including: Decoupling the real-time temperature target offset timing-related feature vector and the heating element power timing-related feature vector to obtain a sequence of real-time temperature target offset timing-related feature local representation vectors and a sequence of heating element power timing-related feature local representation vectors; Calculating the local semantic association interaction vector between each corresponding group of the real-time temperature target offset timing feature local representation vector and the heating element power timing feature local representation vector in the sequence of the real-time temperature target offset timing feature local representation vector and the sequence of the heating element power timing feature local representation vector to obtain a sequence of temperature offset-heating power local semantic association interaction vectors; Calculating the fine-grained attention weight of each temperature offset-heating power local semantic association interaction vector in the sequence of temperature offset-heating power local semantic association interaction vectors to obtain a sequence of fine-grained attention weights; Inputting the sequence of fine-grained attention weights into a random dropout module to obtain a sequence of fine-grained attention weights after dropout processing; The sequence of fine-grained attention weights after the deactivation process is used as the weight distribution, and the weighted sum of the sequence of the temperature offset-heating power local semantic association interaction vectors is calculated to obtain the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

5. The precise control method of the intelligent heating digestion furnace according to claim 4 is characterized in that: Calculating the local semantic association interaction vector between each corresponding group of the real-time temperature target offset timing feature local representation vector and the heating element power timing feature local representation vector in the sequence of the real-time temperature target offset timing feature local representation vector and the sequence of the heating element power timing feature local representation vector to obtain a sequence of temperature offset-heating power local semantic association interaction vectors, including: Calculate the position point addition, position point subtraction and position point multiplication between the local representation vector of the real-time temperature target offset time series feature and the local representation vector of the heating element power time series feature corresponding thereto to obtain a first fusion result, a second fusion result and a third fusion result; After the first fusion result, the second fusion result and the third fusion result are cascaded into a temperature offset-heating power multi-scale fusion feature vector, one-dimensional convolution processing and maximum pooling processing are performed on it to obtain the temperature offset-heating power local semantic association interaction vector.

6. The precise control method of the intelligent heating digestion furnace according to claim 5, characterized in that: Calculating the fine-grained attention weight of each temperature offset-heating power local semantic association interaction vector in the sequence of temperature offset-heating power local semantic association interaction vectors to obtain a sequence of fine-grained attention weights, including: The temperature offset-heating power local semantic association interaction vector is multiplied by a weight coefficient matrix, and then added to the bias vector and passed through a tanh function to obtain a temperature offset-heating power local semantic association interaction encoding feature vector; The attention score reference vector is multiplied by the temperature offset-heating power local semantic association interaction encoding feature vector to obtain the fine-grained attention weight.

7. The precise control method of the intelligent heating digestion furnace according to claim 6, characterized in that: Determining a power adjustment instruction based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector includes: The temperature offset-heating power decoupling enhanced interactive control reasoning representation vector is input into a classifier-based power regulator to obtain the power adjustment instruction, and the power adjustment instruction is used to indicate whether the heating power at the next time point should increase, decrease or remain unchanged.

8. The precise control method of the intelligent heating digestion furnace according to claim 7, characterized in that: Inputting the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector into a classifier-based power regulator to obtain the power adjustment instruction, wherein the power adjustment instruction is used to indicate that the heating power at the next time point should be increased, decreased, or kept unchanged, including: Using multiple fully connected layers of the classifier to fully connect encode the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector to obtain an encoded classification feature vector; The encoded classification feature vector is passed through the Softmax classification function of the classifier to obtain the power adjustment instruction.

9. A precise control system for an intelligent heating digestion furnace, characterized in that: include: The temperature data acquisition module is used to obtain the time queue of the real-time temperature in the digestion furnace collected by the temperature sensor, and set the temperature value; A real-time temperature target offset calculation module, used for calculating the difference between each real-time temperature in the digestion furnace and the set temperature value in the time queue of the real-time temperature in the digestion furnace to obtain a time queue of the real-time temperature target offset; A power data acquisition module, used to acquire a time queue of the real-time power of the heating element collected by the power sensor; A time series variation feature extraction module, used to extract the time series variation features of the time series of the real-time temperature target offset and the time series of the real-time power respectively to obtain a real-time temperature target offset time series correlation feature vector and a heating element power time series correlation feature vector; A local fine-grained interactive response analysis module, used for performing local fine-grained interactive response analysis on the real-time temperature target offset timing correlation feature vector and the heating element power timing correlation feature vector to obtain a temperature offset-heating power decoupling enhanced interactive control reasoning representation vector; The adjustment instruction determination module is used to determine the power adjustment instruction based on the temperature offset-heating power decoupling enhanced interactive control reasoning representation vector.

Images