Automatic control system and method for coal blending and bunker adding

Abstract

The invention discloses an automatic control system and method for coal blending and bunker adding. The method comprises the following steps: firstly, acquiring coal material level values at a plurality of preset time points within a preset time period, which are acquired by a continuous material level indicator deployed in a raw coal bunker, and coal flow values at the plurality of preset time points, which are acquired by a belt weigher deployed on the discharging side of the raw coal bunker; performing time sequence analysis on the coal flow values at the plurality of preset time points and the coal material level values at the plurality of preset time points to obtain a sequence of a coal flow local time sequence feature vector and a coal material level time sequence feature vector; and feature expression optimization is carried out on the coal material level time sequence feature vector to obtain an updated coal material level time sequence feature vector, and finally, whether coal blending and bunker adding are carried out or not is determined based on the updated coal material level time sequence feature vector. Therefore, the coal level of the raw coal bunker can be detected in this way, and whether coal blending and bunker adding are needed or not is judged based on the coal level.

Description

Technical Field

[0001] The present application relates to the field of automated control, and more specifically, to an automated control system and method for coal blending and bunkering. Background Art

[0002] Coal blending refers to the process of mixing coal of different grades in a certain proportion to achieve combustion efficiency and environmental protection requirements. The coal blending scheme based on manual control has the following disadvantages: first, the slow response speed of manual control leads to unreasonable coal blending, which affects the operating efficiency and safety of the boiler; second, the poor accuracy of manual control is prone to human errors, resulting in coal blending deviations, increasing costs and emissions.

[0003] The coal blending and bunkering automation control system is a system that can realize the automatic switching of raw coal bunkers and real-time monitoring of coal material levels according to different bunkering schemes. Coal blending and bunkering automation is a relatively practical technology in the full automation of coal transportation program control. The main difficulty lies in how to effectively detect the coal level in the raw coal bunker. The coal level in the raw coal bunker is a major basis for realizing coal bunker switching.

[0004] Generally speaking, the raw coal bunker is equipped with a continuous level meter, and a belt scale is installed on the discharge side. The level data and coal flow data collected by the continuous level meter and the belt scale can be used as a means of measuring the coal level in the raw coal bunker, but there are certain defects, such as the unstable accuracy of the continuous level meter and the inability to detect bulk coal, while the belt scale measures the process coal volume and cannot be directly converted into the stored coal volume. Therefore, an optimized coal blending and bunkering automatic control system and method are expected. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides an automatic control system and method for coal blending and bunkering, which can use the coal level value collected by the continuous level meter and the coal flow value collected by the belt scale on the discharge side, and combine the intelligent algorithm to perform time series analysis and feature interaction on the two, so as to use the time series change pattern of the coal flow as the inducing information to express, optimize and correct the distribution of the coal level time series characteristics, so as to solve the problem of unstable accuracy of the continuous level meter.

[0006] According to one aspect of the present application, a method for automatically controlling coal blending and storage is provided, which comprises:

[0007] Obtaining coal material level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous material level meter deployed in a raw coal bunker;

[0008] Acquire the coal flow values ​​at the plurality of predetermined time points collected by a belt scale deployed at the discharge side of the raw coal bin;

[0009] Performing time series analysis on the coal flow values ​​at the plurality of predetermined time points and the coal material level values ​​at the plurality of predetermined time points to obtain a sequence of local time series feature vectors of the coal flow and a sequence of time series feature vectors of the coal material level;

[0010] Based on the sequence of the local time series feature vectors of the coal flow, optimizing the feature expression of the coal material level time series feature vector to obtain an updated coal material level time series feature vector; and

[0011] Based on the updated coal material level time series feature vector, it is determined whether to add coal to the warehouse.

[0012] According to another aspect of the present application, there is provided a coal blending and storage automatic control system, which includes:

[0013] A material level value acquisition module is used to acquire coal material level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous material level meter deployed in a raw coal bunker;

[0014] A coal flow value acquisition module, used to acquire the coal flow values ​​at the plurality of predetermined time points collected by a belt scale deployed at the discharge side of the raw coal bin;

[0015] A time series analysis module, used for performing time series analysis on the coal flow values ​​at the plurality of predetermined time points and the coal material level values ​​at the plurality of predetermined time points to obtain a sequence of local time series feature vectors of the coal flow and a time series feature vector of the coal material level;

[0016] A feature expression optimization module, configured to optimize the coal material level time series feature vector based on the sequence of the coal flow local time series feature vectors to obtain an updated coal material level time series feature vector; and

[0017] The control module is used to confirm whether to add coal to the warehouse based on the updated coal material level time series feature vector.

[0018] Compared with the prior art, the automatic control system and method for coal blending and adding to warehouses provided by the present application first obtains the coal level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous level meter deployed in the raw coal bin and the coal flow values ​​at the multiple predetermined time points collected by a belt scale deployed on the discharge side of the raw coal bin, and then performs time series analysis on the coal flow values ​​at the multiple predetermined time points and the coal level values ​​at the multiple predetermined time points to obtain a sequence of local time series feature vectors of coal flow and a coal level time series feature vector, then, based on the sequence of local time series feature vectors of coal flow, performs feature expression optimization on the coal level time series feature vector to obtain an updated coal level time series feature vector, and finally, based on the updated coal level time series feature vector, confirms whether to blend coal and add to warehouses. In this way, the coal level in the raw coal bin can be detected in this way, and based on this, it can be judged whether coal blending and adding to warehouses is needed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0020] Figure 1 The present invention is a flowchart of an automated control method for coal blending and storage according to an embodiment of the present application.

[0021] Figure 2 It is a schematic diagram of the architecture of the automatic control method for coal blending and storage according to an embodiment of the present application.

[0022] Figure 3 This is a flowchart of sub-step S130 of the automatic control method for coal blending and storage according to an embodiment of the present application.

[0023] Figure 4 This is a flowchart of sub-step S150 of the automatic control method for coal blending and storage according to an embodiment of the present application.

[0024] Figure 5 It is a block diagram of an automatic control system for coal blending and storage according to an embodiment of the present application.

[0025] Figure 6 This is an application scenario diagram of the automatic control method for coal blending and storage according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] In response to the above technical problems, the technical concept of this application is to comprehensively utilize the coal level value collected by the continuous level meter and the coal flow value collected by the belt scale on the discharge side, and combine the intelligent algorithm to perform time series analysis and feature interaction on the two, so as to use the time series change pattern of the coal flow as the inducing information to express and optimize the distribution of the coal level time series characteristics and correct and update them, so as to solve the problem of unstable accuracy of the continuous level meter. In this way, the coal level in the raw coal bin is detected, and based on this, it is judged whether coal blending and storage are needed.

[0032] Figure 1 The present invention is a flowchart of an automated control method for coal blending and storage according to an embodiment of the present application. Figure 2 Schematic diagram of the structure of the automatic control method of coal blending and storage according to the embodiment of the present application. Figure 1 and Figure 2 As shown, the automated control method for coal blending and bunkering according to the embodiment of the present application includes the steps of: S110, obtaining coal level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous level meter deployed in a raw coal bin; S120, obtaining coal flow values ​​at the multiple predetermined time points collected by a belt scale deployed on the discharge side of the raw coal bin; S130, performing time series analysis on the coal flow values ​​at the multiple predetermined time points and the coal level values ​​at the multiple predetermined time points to obtain a sequence of local time series feature vectors of coal flow and a coal level time series feature vector; S140, based on the sequence of local time series feature vectors of coal flow, performing feature expression optimization on the coal level time series feature vector to obtain an updated coal level time series feature vector; and, S150, confirming whether to blend coal and bunker it based on the updated coal level time series feature vector.

[0033] Specifically, in the technical solution of the present application, first, the coal level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous level meter deployed in the raw coal bin are obtained; and the coal flow values ​​at the multiple predetermined time points collected by a belt scale deployed on the discharge side of the raw coal bin are obtained.

[0034] Then, after arranging the coal flow values ​​at the multiple predetermined time points into a coal flow time series input vector according to the time dimension, the coal flow time series input vector is vector segmented to obtain a sequence of coal flow local time series input vectors; then, the sequence of coal flow local time series input vectors is passed through a coal flow time series feature extractor based on a one-dimensional convolution layer to obtain a sequence of coal flow local time series feature vectors. Here, vector segmentation of the coal flow time series input vector can divide the coal flow time series input vector into multiple local segments, that is, a sequence of coal flow local time series input vectors, each segment representing a change in coal flow within a shorter time window. In this way, the characteristics of the coal flow can be analyzed more finely within the local time window, and more detailed time series patterns can be captured. The local time series association patterns contained in each coal flow local time series input vector are captured through a one-dimensional convolution layer to understand the change pattern of the coal flow.

[0035] Next, the coal level values ​​at the plurality of predetermined time points are arranged according to the time dimension as a coal level time series input vector, and then the coal level time series encoding feature extractor based on a one-dimensional convolutional layer is used to obtain a coal level time series feature vector. That is, the distribution of coal level time series changes expressed by the coal level values ​​at the plurality of predetermined time points is captured.

[0036] Accordingly, in step S130, Figure 3 As shown, a time series analysis is performed on the coal flow values ​​at the multiple predetermined time points and the coal material level values ​​at the multiple predetermined time points to obtain a sequence of local time series feature vectors of coal flow and a coal material level time series feature vector, including: S131, data preprocessing is performed on the coal flow values ​​at the multiple predetermined time points and the coal material level values ​​at the multiple predetermined time points to obtain a coal flow time series input vector and a coal material level time series input vector; S132, vector segmentation is performed on the coal flow time series input vector to obtain a sequence of local time series input vectors of coal flow; and, S133, time series feature extraction is performed on the sequence of local time series input vectors of coal flow and the coal material level time series input vector using a first deep learning network model and a second deep learning network model to obtain a sequence of local time series feature vectors of coal flow and a coal material level time series feature vector.

[0037] It should be understood that steps S131, S132, and S133 are part of the time series analysis, which is used to obtain the time series feature vector sequence of coal flow and coal material level. In step S131, data preprocessing is performed on the coal flow values ​​and coal material level values ​​at multiple predetermined time points. The purpose of data preprocessing is to clean and convert the original data to make it suitable for subsequent analysis. The preprocessing operation may include data cleaning, missing value processing, outlier processing, data smoothing, etc. Through this step, the time series input vector of coal flow and the time series input vector of coal material level can be obtained. In step S132, the time series input vector of coal flow is segmented to obtain a sequence of local time series input vectors of coal flow. The purpose of vector segmentation is to divide the long time series vector into multiple shorter local time series vectors so as to better capture the local features in the time series data. By segmenting the time series data into a sequence of local time series input vectors, the subsequent time series feature extraction can be better performed. In step S133, the first deep learning network model and the second deep learning network model are used to extract time series features from the sequence of local time series input vectors of coal flow and the time series input vectors of coal material level. The deep learning network model is a powerful tool that can automatically learn features in the data. By using the deep learning network model, useful time series features can be extracted from the time series input vectors of coal flow and coal material level. These time series features can be used to describe the time series change patterns of coal flow and coal material level, providing a basis for subsequent analysis and prediction. In summary, these steps (S131, S132, S133) play an important role in time series analysis, and are used to preprocess, segment and extract features from the time series data of coal flow and coal material level to obtain information about time series features.

[0038] Among them, in step S131, data preprocessing is performed on the coal flow values ​​at the multiple predetermined time points and the coal material level values ​​at the multiple predetermined time points to obtain a coal flow timing input vector and a coal material level timing input vector, including: arranging the coal flow values ​​at the multiple predetermined time points according to the time dimension into the coal flow timing input vector; and arranging the coal material level values ​​at the multiple predetermined time points according to the time dimension into the coal material level timing input vector.

[0039] Wherein, in step S133, the first deep learning network model is a coal flow timing feature extractor based on a one-dimensional convolutional layer, and the second deep learning network model is a material level timing coding feature extractor based on a one-dimensional convolutional layer; wherein, the first deep learning network model and the second deep learning network model are used to perform timing feature extraction on the sequence of the coal flow local timing input vector and the coal material level timing input vector to obtain the sequence of the coal flow local timing feature vector and the coal material level timing feature vector, including: passing the sequence of the coal flow local timing input vector through the coal flow timing feature extractor based on the one-dimensional convolutional layer to obtain the sequence of the coal flow local timing feature vector; and, passing the coal material level timing input vector through the material level timing coding feature extractor based on the one-dimensional convolutional layer to obtain the coal material level timing feature vector.

[0040] In one example, the coal flow time series feature extractor based on the one-dimensional convolution layer includes an input layer, a one-dimensional convolution layer, an activation layer, a pooling layer and an output layer. It is worth mentioning that the one-dimensional convolution layer is a neural network layer commonly used in deep learning, which is used to process one-dimensional sequence data, and it plays an important role in the feature extraction of time series data. The function of the one-dimensional convolution layer is to learn local patterns and features in the input data through convolution operations. It performs convolution operations on the input sequence by sliding a convolution kernel (also called a filter) to extract local features at different positions. The convolution kernel is a small weight matrix that calculates the output value by multiplying and summing the subsequences of the input sequence element by element. By using a one-dimensional convolution layer, local patterns and features in time series data can be effectively captured. The parameter sharing and translation invariance characteristics of the convolution layer enable it to share the learned features at different positions, thereby improving the efficiency and generalization ability of the model. Specifically, the coal flow time series feature extractor based on the one-dimensional convolution layer uses a one-dimensional convolution layer to learn the time series features in the local time series input vector of the coal flow. By performing convolution operations on different positions of the input vector, the convolution layer can extract local features at different positions and output a sequence of local time series feature vectors of coal flow. These time series feature vectors can be used for subsequent analysis and prediction tasks.

[0041] Further, based on the sequence of the local time series feature vector of the coal flow, the coal material level time series feature vector is optimized for feature expression to obtain an updated coal material level time series feature vector. It should be understood that the time series variation characteristics of the coal flow can reflect the state of the coal bunker, such as the vibration state and the stable state. The distribution of the local time series variation characteristics of the coal flow can be used to correct and predict the material level value, so as to better simulate and understand the state of the coal bunker.

[0042] Specifically, when the coal bunker is in a vibrating state, the time series change of the coal flow rate may show obvious fluctuations and instability. When the coal bunker is in a stable state, the time series change of the coal flow rate may show a relatively stable and continuous trend. By simulating the state of the coal bunker and correcting the material level value based on the time series change information of the coal flow rate, the operation of the coal bunker can be understood more accurately.

[0043] In a specific example of the present application, based on the sequence of the local time series feature vectors of the coal flow, the coal material level time series feature vector is optimized for feature expression to obtain an updated coal material level time series feature vector, including: optimizing the coal material level time series feature vector for feature expression using the following feature expression optimization formula to obtain the updated coal material level time series feature vector; wherein the feature expression optimization formula is:

[0044]

[0045] Where A is 1×N w The matrix, N w is the dimension of the coal material level time series feature vector, V is the coal material level time series feature vector, and B is 1×N h The matrix, N h is the dimension of each local time series feature vector of the coal flow, N is the number of local time series feature vectors of the coal flow, h i is the ith coal flow local time series feature vector in the sequence of the coal flow local time series feature vector, σ is the Sigmoid function, V ′ is the updated coal material level time series feature vector, M w and M h Represents the convolution operation of a 1×1 convolution kernel, and S is the weight value.

[0046] Here, by constructing the correlation and importance between each local time series feature vector of coal flow and the time series feature vector of coal material level, the weight values ​​of the feature distribution in different local areas are dynamically allocated to strengthen the features with a high correlation with the state change of the coal bunker, thereby improving the representation ability and expression effect of the coal material level time series feature vector.

[0047] Then, the updated coal material level time series feature vector is passed through a decoder to obtain a decoded value, which is used to represent the updated value of the coal material level value at the current time point; and based on the comparison between the decoded value and a predetermined threshold, it is determined whether to add coal to the warehouse.

[0048] Accordingly, in step S150, Figure 4As shown, based on the updated coal material level timing characteristic vector, confirming whether to add coal to the warehouse comprises: S151, performing characteristic distribution correction on the updated coal material level timing characteristic vector to obtain a corrected updated coal material level timing characteristic vector; S152, passing the corrected updated coal material level timing characteristic vector through a decoder to obtain a decoded value, wherein the decoded value is used to represent an updated value of the coal material level value at the current time point; and, S153, confirming whether to add coal to the warehouse based on a comparison between the decoded value and a predetermined threshold value.

[0049] It should be understood that in step S151, the updated coal material level time series feature vector is subjected to feature distribution correction. The purpose of feature distribution correction is to normalize or standardize the time series feature vector so that it has a uniform distribution characteristic. Through feature distribution correction, the scale difference between different time points can be eliminated, making the feature vector more comparable and consistent. In step S152, the corrected updated coal material level time series feature vector is converted into a decoded value using a decoder. The decoder is a neural network model that maps the time series feature vector to the corresponding updated value of the coal material level value. The decoded value represents the updated value of the coal material level value at the current time point, which can be used to represent the change of the coal material level. In step S153, based on the comparison between the decoded value and the predetermined threshold, it is confirmed whether coal blending and storage are required. The predetermined threshold is a threshold set in advance, which is used to determine whether the coal material level is lower or higher than the expected range. By comparing the decoded value with the predetermined threshold, it can be determined whether the current coal material level needs to be added. If the decoded value exceeds the threshold, it means that the coal level is too low and needs to be restocked; if the decoded value is lower than the threshold, it means that the coal level is within the expected range and does not need to be restocked. In summary, steps S151, S152, and S153 are used to update the time series feature vector of the coal level, and to determine whether it is necessary to perform coal blending and restocking operations based on the comparison between the decoded value and the predetermined threshold. These steps play an important role in the monitoring and control of the coal level, and can judge the changes in the coal level in real time and take corresponding measures.

[0050] In the technical solution of the present application, the coal material level time series feature vector expresses the time series correlation characteristics of the coal material level value, and the sequence of the coal flow local time series feature vector expresses the time series distribution of the coal flow time series correlation characteristics in the local time domain based on vector segmentation under the full time series of the coal flow. Therefore, when the feature expression of the coal material level time series feature vector is optimized based on the sequence of the coal flow local time series feature vector, it is essentially an interpolation dimensional information mixed optimization of the time series correlation characteristics of the coal material level value based on the time series distribution of the coal flow time series correlation characteristics in the local time domain based on vector segmentation under the full time series of the coal flow expressed by the sequence of the coal flow local time series feature vector.

[0051] That is, when the updated coal material level time series feature vector is decoded and regressed through a decoder, the updated coal material level time series feature vector is an interpolated multi-dimensional feature association mixture for the time series distribution decoding regression target relative to the coal material level time series feature vector. In this way, in order to improve the interpolated multi-dimensional feature association mixture enhanced expression effect of the updated coal material level time series feature vector based on the consistency of the sequence feature distribution expression of the local time series feature vector of the coal flow, the updated coal material level time series feature vector is corrected based on the sequence of the local time series feature vector of the coal flow to obtain a corrected feature vector.

[0052] Accordingly, in step S151, in one example, performing characteristic distribution correction on the updated coal material level time series characteristic vector to obtain a corrected updated coal material level time series characteristic vector includes: performing characteristic distribution correction on the updated coal material level time series characteristic vector using the following correction formula to obtain the corrected updated coal material level time series characteristic vector; wherein the correction formula is:

[0053]

[0054] Wherein, V1 is the cascade feature vector obtained by cascading the sequence of the local time series feature vectors of the coal flow, V2 is the updated coal material level time series feature vector, v 1max -1 and v 2max -1 represents the inverse of the global maximum value of the cascade feature vector V1 and the updated coal material level time series feature vector V2, I is a unit vector, and V2 ⊙-1 represents the inverse of the position-by-position characteristic value of the updated coal material level time series characteristic vector V2, ⊙ represents the position point multiplication, represents vector subtraction, ⊕ represents vector addition, V c It is the corrected updated coal material level time series feature vector.

[0055] Specifically, for the correlation interpolation time series association mixing of the regression target in the feature extraction process, based on the idea of ​​interpolation regularization, by unmixing the feature mapping of the outlier feature, the high-dimensional feature manifold is restored to the manifold geometry based on weak enhancement based on the inductive bias, and the consistent feature enhancement mapping of the interpolation samples and interpolation predictions based on feature extraction is realized, so as to obtain the interpolation multi-dimensional feature association mixed enhancement expression effect while maintaining the consistency of the sequence feature distribution expression of the updated coal material level time series feature vector in the local time series feature vector of the coal flow, thereby improving the sequence distribution decoding regression expression effect of the local time series feature vector of the coal flow, so as to improve the accuracy of the decoding result obtained by the decoder of the updated coal material level time series feature vector.

[0056] Further, in step S152, the corrected updated coal material level time series feature vector is passed through a decoder to obtain a decoded value, and the decoded value is used to represent the updated value of the coal material level value at the current time point, including: using multiple fully connected layers of the decoder to decode and regress the corrected updated coal material level time series feature vector using the following decoding formula to obtain the decoded value, wherein the decoding formula is: Among them, V c is the updated coal material level time series feature vector after correction, Y is the decoded value, M1 is the weight matrix, C is the bias vector, represents matrix multiplication and h(·) is the activation function.

[0057] It is worth mentioning that the decoder is a component in deep learning that is used to map the input data from the latent space (usually a low-dimensional representation) back to the original data space. Decoders are often used in models such as autoencoders, variational autoencoders, and generative adversarial networks. The role of the decoder is to convert the latent representation or feature vector into the reconstruction or generation of the original data. It usually consists of one or more neural network layers, which perform feature conversion and reconstruction in an inverse manner. The structure and parameters of the decoder correspond to those of the encoder, and together they constitute a complete autoencoder model. In the image generation task, the decoder can map the vector in the latent space back to the image space to generate realistic images. In the speech synthesis task, the decoder can convert the latent representation into a sound waveform. In the text generation task, the decoder can convert the latent representation into natural language text. The design and training goal of the decoder is to reconstruct or generate the original data as accurately as possible. It captures the structure and pattern of the input data by learning the distribution and characteristics of the data. In other words, the decoder is a component in the deep learning model that is used to map the latent representation back to the original data space, thereby realizing the reconstruction or generation of the data. It plays a key role in autoencoders and generative models.

[0058] In summary, the automated control method for coal blending and adding coal to the warehouse based on the embodiment of the present application is explained, which can detect the coal position in the raw coal bin in this way, and based on this, determine whether coal blending and adding coal to the warehouse are needed.

[0059] Figure 5 FIG. 1 is a block diagram of an automatic control system 100 for coal blending and storage according to an embodiment of the present application. Figure 5As shown, according to the embodiment of the present application, the automatic control system 100 for coal blending and bunkering includes: a material level value acquisition module 110, which is used to obtain the coal level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous level meter deployed in the raw coal bin; a coal flow value acquisition module 120, which is used to obtain the coal flow values ​​at the multiple predetermined time points collected by a belt scale deployed on the discharge side of the raw coal bin; a timing analysis module 130, which is used to perform timing analysis on the coal flow values ​​at the multiple predetermined time points and the coal level values ​​at the multiple predetermined time points to obtain a sequence of local timing feature vectors of coal flow and a coal level timing feature vector; a feature expression optimization module 140, which is used to perform feature expression optimization on the coal level timing feature vector based on the sequence of local timing feature vectors of coal flow to obtain an updated coal level timing feature vector; and a control module 150, which is used to confirm whether to blend coal and bunker based on the updated coal level timing feature vector.

[0060] In one example, in the above-mentioned coal blending and storage automation control system 100, the timing analysis module 130 includes: a data preprocessing unit, which is used to perform data preprocessing on the coal flow values ​​at the multiple predetermined time points and the coal material level values ​​at the multiple predetermined time points to obtain a coal flow timing input vector and a coal material level timing input vector; a vector segmentation unit, which is used to perform vector segmentation on the coal flow timing input vector to obtain a sequence of coal flow local timing input vectors; and a timing feature extraction unit, which is used to perform timing feature extraction on the sequence of coal flow local timing input vectors and the coal material level timing input vector using the first deep learning network model and the second deep learning network model to obtain a sequence of coal flow local timing feature vectors and the coal material level timing feature vector.

[0061] Here, those skilled in the art will appreciate that the specific functions and operations of each module in the above-mentioned coal blending and bunkering automatic control system 100 have been described in detail above. Figures 1 to 4 The description of the automatic control method for coal blending and bunkering has been introduced in detail, and therefore, its repeated description will be omitted.

[0062] As described above, the coal blending and storage automation control system 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a coal blending and storage automation control algorithm. In one example, the coal blending and storage automation control system 100 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 coal blending and storage automation control system 100 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 coal blending and storage automation control system 100 can also be one of the many hardware modules of the wireless terminal.

[0063] Alternatively, in another example, the coal blending and storage automation control system 100 and the wireless terminal may be separate devices, and the coal blending and storage automation control system 100 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.

[0064] Figure 6 FIG. 1 is an application scenario diagram of the automatic control method for coal blending and storage according to an embodiment of the present application. Figure 6 As shown, in this application scenario, first, the coal material level values ​​(for example, Figure 6 D1) and the coal flow values ​​at the plurality of predetermined time points collected by the belt scale deployed at the discharge side of the raw coal bunker (for example, Figure 6 Then, the coal flow values ​​at the plurality of predetermined time points and the coal material level values ​​at the plurality of predetermined time points are input to a server (for example, Figure 6 In S) shown in , the server can use the coal blending and storage automation control algorithm to process the coal flow values ​​and the coal level values ​​at the multiple predetermined time points to confirm whether to blend and store coal.

[0065] It should be understood that the automation of coal blending and bunkering is mainly to solve the problem of how to realize different bunkering solutions. The automation of coal blending and bunkering is a relatively practical technology in the full automation of coal transportation program control. Its theoretical difficulty is relatively low, but its actual implementation is relatively difficult. The main difficulty lies in how to effectively detect the coal position in the raw coal bunker. The coal position in the raw coal bunker is a major basis for realizing coal bunker switching.

[0066] Generally speaking, the raw coal bunker is equipped with continuous material level meter and material level switch, and the feeding side of the raw coal bunker is equipped with upper bunker belt scale, and the discharging side is equipped with coal feeding belt scale. The material level meter and the coal inlet and outlet belt scale can be used as the means of coal level measurement in the raw coal bunker, but both have certain defects, such as the material level meter has unstable accuracy and cannot detect bulk coal, and the belt scale measures the process coal volume and cannot be directly converted into the stored coal volume (no initial value).

[0067] The raw coal bunker material level measurement method that meets practical applications can not only be used for coal blending and bunkering, but also has many other uses. For example, it can detect whether the continuous material level meter is faulty and whether the coal is loose, so as to further decide whether to automatically start the air cannon vibration, thereby optimizing the operation of the air cannon.

[0068] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0069] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0070] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application are described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A method for automatic control of coal blending and storage, characterized in that: include: Obtaining coal material level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous material level meter deployed in a raw coal bunker; Acquire the coal flow values ​​at the plurality of predetermined time points collected by a belt scale deployed at the discharge side of the raw coal bin; Performing time series analysis on the coal flow values ​​at the plurality of predetermined time points and the coal material level values ​​at the plurality of predetermined time points to obtain a sequence of local time series feature vectors of the coal flow and a sequence of time series feature vectors of the coal material level; Based on the sequence of the local time series feature vectors of the coal flow, optimizing the feature expression of the coal material level time series feature vector to obtain an updated coal material level time series feature vector; as well as Based on the updated coal material level time series feature vector, it is determined whether to add coal to the warehouse.

2. The automatic control method for coal blending and storage according to claim 1, characterized in that: Performing time series analysis on the coal flow values ​​at the plurality of predetermined time points and the coal material level values ​​at the plurality of predetermined time points to obtain a sequence of local time series feature vectors of coal flow and a time series feature vector of coal material level, including: Performing data preprocessing on the coal flow values ​​at the plurality of predetermined time points and the coal material level values ​​at the plurality of predetermined time points to obtain a coal flow time series input vector and a coal material level time series input vector; Performing vector segmentation on the coal flow time series input vector to obtain a sequence of coal flow local time series input vectors; and The first deep learning network model and the second deep learning network model are used to perform time series feature extraction on the sequence of the coal flow local time series input vector and the coal material level time series input vector to obtain the sequence of the coal flow local time series feature vector and the coal material level time series feature vector.

3. The automatic control method for coal blending and storage according to claim 2 is characterized in that: The coal flow rate values ​​and the coal material level values ​​at the plurality of predetermined time points are subjected to data preprocessing to obtain a coal flow rate time series input vector and a coal material level time series input vector, including: Arranging the coal flow values ​​at the plurality of predetermined time points according to the time dimension into the coal flow time series input vector; and The coal level values ​​at the plurality of predetermined time points are arranged according to the time dimension to form the coal level time series input vector.

4. The automatic control method for coal blending and storage according to claim 3 is characterized in that: The first deep learning network model is a coal flow time series feature extractor based on a one-dimensional convolutional layer, and the second deep learning network model is a material level time series coding feature extractor based on a one-dimensional convolutional layer; Among them, using the first deep learning network model and the second deep learning network model to extract the time series features of the sequence of the coal flow local time series input vector and the coal material level time series input vector to obtain the sequence of the coal flow local time series feature vector and the coal material level time series feature vector, including: Passing the sequence of the coal flow local time series input vectors through the coal flow time series feature extractor based on the one-dimensional convolution layer to obtain the sequence of the coal flow local time series feature vectors; and The coal material level time series input vector is passed through the material level time series coding feature extractor based on the one-dimensional convolution layer to obtain the coal material level time series feature vector.

5. The automatic control method for coal blending and storage according to claim 4, characterized in that: The coal flow time series feature extractor based on one-dimensional convolution layer includes an input layer, a one-dimensional convolution layer, an activation layer, a pooling layer and an output layer.

6. The automatic control method for coal blending and storage according to claim 5, characterized in that: Based on the sequence of the local time series feature vectors of the coal flow, the feature expression of the coal material level time series feature vector is optimized to obtain an updated coal material level time series feature vector, including: The coal material level time series feature vector is optimized by the following feature expression optimization formula to obtain the updated coal material level time series feature vector; Among them, the feature expression optimization formula is: Where A is 1×N w The matrix, N w is the dimension of the coal material level time series feature vector, V is the coal material level time series feature vector, and B is 1×N h The matrix, N h is the dimension of each local time series feature vector of the coal flow, N is the number of local time series feature vectors of the coal flow, h i is the ith coal flow local time series feature vector in the sequence of the coal flow local time series feature vectors, σ is the Sigmoid function, V′ is the updated coal material level time series feature vector, M w and M h Represents the convolution operation of a 1×1 convolution kernel, and S is the weight value.

7. The automatic control method for coal blending and storage according to claim 6, characterized in that: Based on the updated coal material level time series feature vector, determining whether to add coal to the warehouse includes: Performing characteristic distribution correction on the updated coal material level time series characteristic vector to obtain a corrected updated coal material level time series characteristic vector; Passing the corrected updated coal material level time series feature vector through a decoder to obtain a decoded value, wherein the decoded value is used to represent an updated value of the coal material level value at the current time point; and Based on the comparison between the decoded value and a predetermined threshold, it is determined whether to add coal to the warehouse.

8. The automatic control method for coal blending and storage according to claim 7, characterized in that: The corrected updated coal material level time series feature vector is passed through a decoder to obtain a decoded value, and the decoded value is used to represent the updated value of the coal material level value at the current time point, including: The multiple fully connected layers of the decoder are used to perform decoding regression on the corrected updated coal material level time series feature vector using the following decoding formula to obtain the decoding value, wherein the decoding formula is: Among them, V c is the updated coal material level time series feature vector after correction, Y is the decoded value, M1 is the weight matrix, C is the bias vector, represents matrix multiplication and h(·) is the activation function.

9. An automatic control system for coal blending and storage, characterized in that: include: A material level value acquisition module is used to acquire coal material level values ​​at multiple predetermined time points within a predetermined time period collected by a continuous material level meter deployed in a raw coal bunker; A coal flow value acquisition module, used to acquire the coal flow values ​​at the plurality of predetermined time points collected by a belt scale deployed at the discharge side of the raw coal bin; A time series analysis module, used for performing time series analysis on the coal flow values ​​at the plurality of predetermined time points and the coal material level values ​​at the plurality of predetermined time points to obtain a sequence of local time series feature vectors of the coal flow and a time series feature vector of the coal material level; A feature expression optimization module, configured to optimize the feature expression of the coal material level time series feature vector based on the sequence of the coal flow local time series feature vector to obtain an updated coal material level time series feature vector; as well as The control module is used to confirm whether to add coal to the warehouse based on the updated coal material level time series feature vector.

10. The automatic control system for coal blending and storage according to claim 9, characterized in that: The timing analysis module comprises: A data preprocessing unit, configured to perform data preprocessing on the coal flow values ​​and the coal material level values ​​at the plurality of predetermined time points to obtain a coal flow time series input vector and a coal material level time series input vector; a vector segmentation unit, configured to perform vector segmentation on the coal flow time series input vector to obtain a sequence of coal flow local time series input vectors; and A timing feature extraction unit is used to use the first deep learning network model and the second deep learning network model to perform timing feature extraction on the sequence of the local timing input vector of the coal flow and the timing input vector of the coal material level to obtain the sequence of the local timing feature vector of the coal flow and the timing feature vector of the coal material level.

Images