Intelligent industrial energy consumption analysis and control decision system and method

Abstract

The invention discloses an intelligent industrial energy consumption analysis and control decision-making system and method, and relates to the technical field of industrial energy consumption management, and the system comprises an energy consumption information transmission and storage module, an energy consumption information processing and analysis module, and an energy consumption control decision-making module. The energy consumption information transmission and storage module comprises an information transmission unit and an information storage unit; the energy consumption information processing and analyzing module comprises an information processing unit and an information analyzing unit; the energy consumption management and control decision module comprises an energy consumption prediction unit and an energy consumption management and control unit. According to the invention, by using the congestion control delay optimization algorithm, the information transmission rate and the network stability are improved; information processing is performed by using a DPTCN model in the deep learning network, so that the accuracy, efficiency and robustness of information processing are improved; the energy consumption control intelligent decision is performed by using the simulated annealing particle swarm optimization, so that the decision quality and effect are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial energy consumption management, and in particular to an intelligent industrial energy consumption analysis and control decision-making system and method. Background Art

[0002] With the continuous growth of energy demand and the transformation of energy structure, power data has great potential in industrial economic monitoring and energy efficiency diagnosis. The development of modern industrial production and urban infrastructure has led to an increasing trend in energy consumption worldwide, but it has also brought about energy waste and environmental pollution problems. Therefore, how to effectively manage and optimize energy use has become a major task facing governments and enterprises around the world. In this context, the collection, analysis and utilization of power data have gradually become an important means to improve energy efficiency, reduce costs and achieve sustainable development. Through accurate energy monitoring systems, the energy consumption of various equipment and systems can be fully understood, and deficiencies in energy utilization can be discovered through data analysis, and then optimized and adjusted. In addition, with the development of big data, cloud computing and Internet of Things technologies, more and more intelligent energy consumption management systems are being applied to industrial, commercial and residential fields, realizing real-time collection, remote monitoring and intelligent decision support of power data. These systems not only improve energy efficiency but also provide guarantees for the sustainable use of energy.

[0003] However, the existing industrial energy consumption monitoring and control systems still have shortcomings in practical applications. First, untimely data collection and insufficient data accuracy are the main problems facing the current system. Many traditional systems rely on manual operations and regular inspections, resulting in insufficient and incomplete data acquisition, which affects the accuracy of subsequent analysis. Second, existing monitoring systems often lack real-time early warning functions, making it difficult to provide effective warnings and intervention measures before problems occur. This makes it difficult to detect some potential energy efficiency problems and equipment failures in a timely manner, and even leads to premature equipment loss or large-scale energy waste. In addition, existing monitoring methods are mostly simple quantitative statistics, lacking intelligent analysis and prediction functions, and unable to make dynamic adjustments based on multi-dimensional data. Finally, the lack of data processing and analysis capabilities, especially in massive data processing and deep mining, further restricts the intelligent development of energy consumption management systems. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to provide an intelligent industrial energy consumption analysis and control decision-making system to overcome the limitations of traditional energy monitoring and control systems, solve the problems of slow information transmission and storage, information loss, low information processing efficiency and low prediction accuracy in existing industrial energy consumption analysis and control decision-making systems, and further improve the accuracy and stability of energy consumption information transmission and processing efficiency.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides an intelligent industrial energy consumption analysis and control decision-making system, which includes an energy consumption information transmission and storage module, an energy consumption information processing and analysis module and an energy consumption control decision-making module; the energy consumption information transmission and storage module includes an information transmission unit for transmitting energy consumption information; an information storage unit for storing energy consumption information; the energy consumption information processing and analysis module includes an information processing unit for extracting features and converting information from energy consumption information; an information analysis unit for analyzing energy consumption information and optimizing analysis results; the energy consumption control decision-making module includes an energy consumption prediction unit for establishing a prediction model using energy consumption information to predict energy consumption trends; and an energy consumption control unit for formulating energy consumption management strategies and decisions based on energy consumption trends.

[0008] As a preferred solution of the intelligent industrial energy consumption analysis and management decision-making system described in the present invention, the transmission of energy consumption information refers to the efficient transmission of information using the congestion control algorithm in the delay optimization algorithm; the use of the congestion control algorithm to transmit information includes determining information transmission path parameters, calculating delay-tolerant network bandwidth, calculating information transmission rate, and optimizing congestion status.

[0009] As an optimal solution for the intelligent industrial energy consumption analysis and control decision-making system described in the present invention, the storage of energy consumption information refers to the use of HDFS distributed storage technology to efficiently store information, including file segmentation operations, metadata management operations, and information access and operations.

[0010] As an optimal solution for the intelligent industrial energy consumption analysis and control decision-making system described in the present invention, the feature extraction and information conversion of energy consumption information refers to the use of the DPTCN model in the deep learning network to perform key feature extraction and information conversion in information processing tasks; the DPTCN model is composed of a conversion layer, a squeeze layer, an excitation layer, and a recalibration layer.

[0011] As a preferred solution of the intelligent industrial energy consumption analysis and control decision-making system described in the present invention, the following steps are included: analyzing the energy consumption information and optimizing the analysis results, which refers to identifying information patterns and rules using information mining technology, and optimizing the analysis results using reinforcement learning; identifying information patterns and rules using information mining technology refers to discovering hidden patterns and rules from energy consumption information using information mining technology, providing decision-making basis for energy consumption management and operation, and applying association rules to mine the correlation between energy consumption use and industry; optimizing the analysis results using reinforcement learning refers to defining the state space and action space based on actual energy consumption information, including the following steps: training and optimizing using a reinforcement learning algorithm based on the defined environmental model, state space, action space, and reward function; continuously updating the strategy and adjusting the action selection strategy through interaction with the environment; evaluating the reliability of the analysis results through comparison with the real environment or simulation, and adjusting and improving the analysis results.

[0012] As a preferred solution of the intelligent industrial energy consumption analysis and control decision-making system described in the present invention, the use of energy consumption information to establish a prediction model refers to the energy consumption prediction unit using the processed energy consumption information to establish a prediction model to predict future energy consumption trends, so that regulators can make corresponding adjustments and decisions; the prediction of energy consumption trends includes the following steps: extracting the features required for energy consumption prediction according to actual conditions; building an LSTM model using a deep learning framework; training the LSTM model using a training set, and optimizing the model parameters through the back propagation algorithm and loss function during the training process; evaluating the performance of the trained LSTM model using a test set, and calculating the error index between the prediction results and the actual energy consumption values; predicting future energy consumption based on the trained LSTM model, converting the features into an input format acceptable to the model, and using the model to generate corresponding energy consumption prediction results; optimizing and adjusting the parameters of the LSTM model based on the prediction results and feedback from the performance evaluation.

[0013] As a preferred solution of the intelligent industrial energy consumption analysis and control decision-making system described in the present invention, the energy consumption management strategy and decision-making based on energy consumption trends refers to making intelligent energy consumption control decisions based on the simulated annealing particle swarm algorithm; the energy consumption control intelligent decision-making based on the simulated annealing particle swarm algorithm includes the following steps: taking the optimal solution of the PSO algorithm as the initial value, setting the initial annealing temperature, wherein the initial position and speed of the factor are evaluated and compared according to the objective function; calculating the fitness function difference between the new solution and the old solution, and judging whether to accept the new solution according to the Metropolis judgment standard; calculating the annealing temperature; obtaining a new energy consumption plan based on the updated position; introducing the idea of ​​the simulated annealing algorithm on the basis of energy consumption decision-making, and avoiding falling into the local optimal solution by accepting the difference solution with a certain probability; terminating the algorithm according to the set termination condition and outputting the optimal energy consumption plan and the corresponding energy consumption value.

[0014] Secondly, in order to further solve the safety problems existing in industrial energy consumption management, the embodiments of the present invention provide an intelligent industrial energy consumption analysis and control decision-making method, which includes: obtaining energy consumption information, and transmitting and storing the energy consumption information; performing feature extraction and information conversion on the energy consumption information to obtain processed energy consumption information; performing analysis based on the energy consumption information and optimizing the analysis results; using the processed energy consumption information to establish a prediction model to predict energy consumption trends; evaluating the prediction results based on the energy consumption trends, and formulating energy consumption management strategies and decisions.

[0015] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the intelligent industrial energy consumption analysis and control decision-making system as described in the first aspect of the present invention is implemented.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored, wherein: when the computer program is executed by a processor, any step of the intelligent industrial energy consumption analysis and control decision-making system as described in the first aspect of the present invention is implemented.

[0017] The beneficial effects of the present invention are as follows: the present invention improves the information transmission rate and network stability by utilizing a congestion control delay optimization algorithm; utilizes the DPTCN model in a deep learning network for information processing, thereby improving the accuracy, efficiency, and robustness of information processing; and utilizes a simulated annealing particle swarm algorithm for intelligent decision-making on energy consumption control, thereby improving the quality and effectiveness of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0019] Figure 1 This is a schematic diagram of an intelligent industrial energy consumption analysis and control decision-making system in Example 1.

[0020] Figure 2 This is a structural diagram of the computer equipment in Example 3. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0024] Example 1

[0025] Reference Figure 1 , which is the first embodiment of the present invention, provides an intelligent industrial energy consumption analysis and control decision-making system, including an energy consumption information transmission and storage module, an energy consumption information processing and analysis module and an energy consumption control decision-making module.

[0026] Preferably, Figure 1 The figure shows a schematic diagram of an intelligent industrial energy consumption analysis and control decision-making system. The energy consumption information transmission and storage module includes an information transmission unit and an information storage unit, wherein the information transmission unit is used to transmit energy consumption information, and the information storage unit is used to store energy consumption information.

[0027] It should be noted that the information transmission unit is based on network communication technology to build a multi-level, highly reliable information transmission channel to ensure that energy consumption information is transmitted to the information storage unit in real time, accurately and securely, facilitating subsequent timely and stable information management.

[0028] Furthermore, transmitting energy consumption information refers to using the congestion control algorithm in the delay optimization algorithm to efficiently transmit information. The congestion control algorithm is an important part of the transmission control protocol, which is used to avoid network congestion and improve network transmission efficiency. It is achieved by controlling the sending rate of the sender, which can prevent network congestion in the early stage of connection establishment and improve the network's carrying capacity.

[0029] Specifically, using the congestion control algorithm for information transmission includes determining information transmission path parameters, calculating delay-tolerant network bandwidth, calculating information transmission rate, and optimizing congestion status.

[0030] Specifically, determining the information transmission path parameters refers to setting the network parameters that need to be transmitted in the delay-tolerant network. The specific formula is as follows:

[0031]

[0032] Among them, f i is the network parameter that needs to be transmitted in the delay-tolerant network; N is the total amount of information to be transmitted; a i and b i are information set and path set respectively; M is the number of transmission nodes; L is the network transmission path bandwidth; S is the number of network control terminals.

[0033] Specifically, calculating the bandwidth of the delay-tolerant network refers to calculating the bandwidth of the delay-tolerant network based on the feedback speed of the sent signal. The specific formula is as follows:

[0034]

[0035] Where D is the bandwidth of the delay-tolerant network; Q is the delay-tolerant network packet; U j+1 The time it takes for the j+1th group of information to be transmitted to the control terminal; Δ j is the difference between the time when the j+1th group of information is transmitted to the control terminal and the time when the jth group of information is transmitted to the control terminal.

[0036] Specifically, information transmission rate calculation refers to calculating the information transmission rate based on the increase and decrease of the congestion window. The specific formula is as follows:

[0037]

[0038] Where C is the information transmission rate; SUU is the transmission path congestion time; U0 is the delay-tolerant network timeout time; c is the amount of information to be transmitted; and q is the delay-tolerant network group.

[0039] Specifically, congestion state optimization refers to performing congestion state optimization processing to achieve congestion avoidance optimization. The specific formula is as follows:

[0040]

[0041] in, is the congestion state prediction function; y j is the network state information; l is the number of network layers; ε is the activation function; and x jk is the congestion state weight coefficient, where j = 1, 2, ..., P1, k = 1, 2, ..., P2; θ jk and θ2 are bias values.

[0042] Preferably, transmission performance and throughput are improved by applying the congestion control algorithm in the delay optimization algorithm, wherein the congestion control algorithm can optimize network performance and throughput by dynamically adjusting the sending rate; when the network load is light, the sending rate can be gradually increased to fully utilize the available bandwidth; and when the network load is heavy, the sending rate will automatically decrease to avoid congestion and information loss, thereby ensuring stable and efficient transmission operation.

[0043] It should be noted that the information storage unit is responsible for storing various types of information transmitted by the information transmission unit, providing a centralized and reliable storage medium for saving and managing various energy consumption information, including total electricity consumption, consumption trends, upstream and downstream electricity consumption of key industries, and electricity consumption of key enterprises. The information storage unit plays the role of an information warehouse in the system to ensure the persistence and reliability of information.

[0044] Furthermore, storing energy consumption information means using HDFS distributed storage technology to efficiently store information. HDFS distributed storage technology is an open source implementation of Google File System. It is a highly scalable distributed storage technology suitable for storing large-scale unstructured information. HDFS distributed storage technology can store information in a distributed manner on multiple nodes and provide a unified access interface. Users can operate files in HDFS distributed storage technology just like operating a local file system.

[0045] Specifically, using HDFS distributed storage technology for efficient information storage includes the following steps: performing file segmentation operations. When a file is to be stored in HDFS distributed storage technology, it will be divided into fixed-size information blocks (the default is 128MB). These information blocks will be stored in multiple DataNodes in the HDFS cluster. The segmentation and distribution of information blocks are to achieve parallel processing and fault tolerance of information.

[0046] To perform metadata management operations, HDFS distributed storage technology uses a centralized component called NameNode to manage the metadata of the file system, including the structure of files and directories, the location information of file blocks, and access control lists.

[0047] Access and operate information. When reading information, the client obtains the information from the nearest DataNode and can improve reading performance by reading multiple information blocks in parallel. When writing information, the client divides the information blocks into information packets and sends the information packets to multiple DataNodes. The DataNode stores the information packets on the local disk and copies them to other replicas.

[0048] Preferably, the present invention has significant advantages by applying HDFS distributed storage technology to information storage units. First, HDFS distributed storage technology is designed for processing large-scale information sets and can be horizontally expanded. It can store and process PB-level information on thousands of nodes. It adopts a distributed storage and parallel processing mechanism, which can efficiently process large amounts of information and provide high-throughput information access; secondly, HDFS distributed storage technology can expand storage capacity and performance by adding nodes. The storage capacity and performance of HDFS distributed storage technology can grow almost linearly, which can meet the needs of large-scale information storage; finally, HDFS distributed storage technology supports sequential read and write operations and is suitable for batch processing scenarios of large-scale information. Since information blocks are divided and distributed on multiple nodes, parallel read and write operations can provide higher performance.

[0049] Preferably, the energy consumption information processing and analysis module includes an information processing unit and an information analysis unit, wherein the information processing unit is used to extract features and convert information from energy consumption information, and the information analysis unit is used to analyze the energy consumption information and optimize the analysis results.

[0050] It should be noted that the information processing unit is designed to process and analyze the information of the information storage unit, providing essential support for the subsequent information analysis of the system. The role of the information processing unit is to process, organize and convert the energy consumption information in the information storage unit. It is the core of the system operation and an important manifestation of the system's intelligence.

[0051] Furthermore, feature extraction and information conversion of energy consumption information refers to using the DPTCN model in the deep learning network to perform key feature extraction and information conversion in information processing tasks. The DPTCN model is a deep learning model based on convolutional neural networks and attention mechanisms. It consists of a conversion layer, a squeeze layer, an excitation layer, and a recalibration layer, and processes information in an orderly and efficient manner through different layers. In addition, the DPTCN model can effectively extract local and global features from information, and uses the attention mechanism to help the model focus on the important parts of the information. The DPTCN model has strong generalization capabilities and can be applied to the processing of multiple information categories.

[0052] Specifically, the conversion layer refers to the feature extraction of the input through convolution calculation, in which the input information is subjected to standard convolution calculation in the conversion layer to extract the features of local information. The specific formula is as follows:

[0053]

[0054] Among them, u c is the feature map after convolution calculation; x s is the original information set; ν c is the cth convolution kernel, the number of feature channels is C; X is the multi-channel word vector matrix of the input data.

[0055] Furthermore, the convolution calculation of the conversion layer adopts temporal convolution, and the specific formula of the feature channel output is as follows:

[0056] C' i =f(K j ·X i +b i )

[0057] Among them, C' i is the feature channel output at the i-th moment; X i K is the word vector matrix calculated at the convolutional layer at the i-th moment; j is the convolution kernel of the jth layer; b i is the bias vector; f is the nonlinear activation (convolution) function.

[0058] Furthermore, the conversion layer performs standard convolution calculations on the input information to extract the features of local information. The specific formula for extracting the features is as follows:

[0059]

[0060] Among them, Z is the feature finally extracted by the conversion layer; It is the mth maximum pooling feature extraction.

[0061] Specifically, the squeeze layer is to achieve feature compression of the output of the conversion layer through global average pooling (GAP), compressing each two-dimensional feature into a one-dimensional real number with a global sense of the feature channel, making up for the defect of losing context information in the convolution calculation. The specific formula is as follows:

[0062]

[0063] Among them, H is the height of the original feature map; W is the width of the original feature map; C is the number of channels; u c is the original feature map; z c is the compressed feature map.

[0064] Specifically, the incentive layer refers to generating weights for feature channels through parameter training. The specific formula is as follows:

[0065] S=σ(W2δ(W1z))

[0066] Among them, S is the generated weight; W1 is the channel reduction calculation coefficient; W2 is the dimension recovery coefficient; σ is the Sigmoid function; δ is the ReLU activation function; z is the compressed feature map of the input.

[0067] Specifically, the recalibration layer refers to the recalibration of the original features in the feature channel dimension. The specific formula is as follows:

[0068]

[0069] Among them, s c is the activation value, ranging from 0 to 1; u c is the original feature map; Recalibrated output.

[0070] Preferably, the present invention constructs an information processing unit by utilizing the DPTCN model, wherein the structure of the DPTCN model is relatively simple and easy to train and deploy. Compared with other Transformer-based models, the DPTCN model has fewer parameters and faster training and processing. The DPTCN model can be used in combination with other technologies, such as pre-trained models or integrated learning, to further improve information processing performance. The DPTCN model can effectively extract information features and utilize these features for information processing, thereby improving the accuracy of information processing. The DPTCN model can process information in parallel, thereby improving the efficiency of information processing. The DPTCN model can resist noise and interference, thereby improving the robustness of information processing.

[0071] It should be noted that the information analysis unit aims to identify patterns, associations and potential anomalies in energy consumption information by applying information mining and reinforcement learning techniques, and then support the generation and optimization of energy consumption management and control decisions based on the analysis results of energy consumption information.

[0072] Furthermore, analyzing the energy consumption information and optimizing the analysis results refers to using information mining technology to identify information patterns and regularities, and using reinforcement learning to optimize the analysis results.

[0073] Specifically, using information mining technology to identify information patterns and laws means that information mining technology can discover hidden patterns and laws from a large amount of energy consumption information, provide important decision-making basis for energy consumption management and operation, and apply association rules to mine and discover the correlation between energy consumption and industry.

[0074] Specifically, using reinforcement learning to optimize analysis results refers to defining a state space and an action space based on actual energy consumption information, where the state space is the set of all possible states, and the action space is the set of all possible actions. This includes the following steps: using a reinforcement learning algorithm for training and optimization based on the defined environment model, state space, action space, and reward function.

[0075] Through interaction with the environment, the reinforcement learning algorithm continuously updates the strategy and adjusts the action selection strategy to maximize the cumulative reward.

[0076] By comparing with the real environment or through simulation, the reliability of the analysis results can be evaluated, and the analysis results can be adjusted and improved.

[0077] Preferably, the present invention can improve the accuracy and efficiency of information analysis by constructing an information analysis unit using information mining and reinforcement learning technologies. Information mining technology can automatically discover hidden patterns and rules from massive information, helping the information analysis unit to quickly and accurately extract valuable information; reinforcement learning technology can enable the information analysis unit to learn optimal strategies by interacting with the environment, and continuously improve the accuracy and efficiency of information analysis; secondly, it can enhance the depth and breadth of information analysis, among which information mining technology can analyze information from different angles and discover knowledge and information that is difficult to discover with traditional methods, and reinforcement learning technology can take the complexity of the environment into consideration in information analysis, making information analysis closer to the actual situation.

[0078] Preferably, the energy consumption control decision module includes an energy consumption prediction unit and an energy consumption control unit, wherein the energy consumption prediction unit is used to use energy consumption information to establish a prediction model and predict energy consumption trends, and the energy consumption control unit is used to formulate energy consumption management strategies and decisions based on energy consumption trends.

[0079] Furthermore, using energy consumption information to establish a prediction model means that the energy consumption prediction unit uses the processed energy consumption information to establish a prediction model to predict future energy consumption trends, so that regulators can make corresponding adjustments and decisions to achieve reasonable allocation and use of energy resources. Among them, the long short-term memory network LSTM is used for energy consumption prediction because the long short-term memory network LSTM can process sequence data and performs well in time series prediction tasks.

[0080] Specifically, predicting energy consumption trends includes the following steps: extracting features required for energy consumption prediction based on actual conditions.

[0081] Use an appropriate deep learning framework, such as TensorFlow or Keras, to build an LSTM model, where an LSTM model typically consists of one or more LSTM layers and an output layer.

[0082] The LSTM model is trained using the training set. During the training process, the model parameters are optimized through the back-propagation algorithm and the appropriate loss function, so that the model can learn and fit the historical energy consumption series.

[0083] The performance of the trained LSTM model is evaluated using the test set, and the error indicators between the predicted results and the actual energy consumption values ​​are calculated. The error indicators include the root mean square error (RMSE) and the mean absolute percentage error (MAPE) to evaluate the accuracy and generalization ability of the model.

[0084] Based on the trained LSTM model, the future energy consumption is predicted, the features are converted into an input format acceptable to the model, and the model is used to generate the corresponding energy consumption prediction results.

[0085] The LSTM model is optimized and adjusted based on the prediction results and performance evaluation feedback. Different network structures, hyperparameter settings, and regularization methods can be tried to improve the model's predictive ability and stability.

[0086] It should be noted that the energy consumption control unit helps regulators formulate energy consumption management strategies and decisions by providing data support and decision-making recommendations; by analyzing energy consumption data and trends, it provides decision-making support for managers and promotes energy conservation, environmental protection and sustainable development.

[0087] Furthermore, formulating energy consumption management strategies and decisions based on energy consumption trends refers to making intelligent energy consumption control decisions based on the simulated annealing particle swarm algorithm. Simulated annealing (SA) is a heuristic optimization algorithm used to find the optimal solution in the search space. It is derived from the annealing process of solid materials and performs optimization search by simulating the movement process of atoms when solid materials are cooled.

[0088] It should be noted that, since the simulated annealing method SA can accept inferior solutions with a certain probability, the PSO algorithm is prone to falling into the local optimal problem. The present invention solves this problem by introducing the simulated annealing idea into the PSO algorithm.

[0089] Specifically, the intelligent decision-making of energy consumption control based on simulated annealing particle swarm algorithm includes the following steps: taking the optimal solution of the PSO algorithm as the initial value, setting the initial annealing temperature T0, wherein the initial position and speed of the factor are evaluated and compared according to the objective function, and when initializing the factor group, it is necessary to ensure that the generated factors meet the constraints.

[0090] Calculate the fitness function difference Δf = f(Y′) - f(Y) between the new solution and the old solution, and determine whether to accept the new solution according to the Metropolis criterion. The specific formula of the Metropolis criterion is as follows:

[0091]

[0092] Among them, Δf is the difference in fitness function between the new solution and the old solution; T is the temperature parameter, which controls the probability of accepting a worse solution.

[0093] Calculate the annealing temperature. The specific formula is as follows:

[0094] T t =CT0

[0095] Among them, T t is the temperature at time t; C is the cooling rate; T0 is the initial temperature.

[0096] A new energy consumption plan is obtained based on the updated position, where this process is equivalent to the development process in the particle swarm algorithm. By searching the surrounding solution space to find a better energy consumption plan, a local search algorithm can be used for optimization, such as gradient descent.

[0097] Based on the energy consumption decision, the idea of ​​simulated annealing algorithm is introduced to avoid falling into the local optimal solution by accepting the difference solution with a certain probability. The decision of whether to accept the new solution is made based on the energy consumption value and temperature of the current solution. As the iteration proceeds, the temperature gradually decreases and the probability of accepting the difference solution also gradually decreases, thus approaching the global optimal solution.

[0098] The algorithm is terminated according to the set termination conditions and the optimal energy consumption plan and corresponding energy consumption value are output. The termination condition is to reach the maximum number of iterations or meet certain convergence conditions. This optimal energy consumption plan can be used as the result of intelligent decision-making for energy consumption control to guide actual energy consumption management and operation.

[0099] Preferably, the present invention uses simulated annealing particle swarm algorithm to make intelligent decisions on energy consumption control, which can better find the global optimal solution in the energy consumption control problem, avoid falling into the local optimal solution, and improve the quality and effect of decision-making; it can flexibly adjust the search strategy and parameter settings according to actual conditions to adapt to different decision-making scenarios; it has multi-objective optimization capabilities, can balance and optimize among multiple objectives, find a set of non-inferior solutions, and then provide decision makers with multiple feasible options.

[0100] In summary, the present invention improves the information transmission rate and network stability by utilizing the congestion control delay optimization algorithm; uses the DPTCN model in the deep learning network for information processing, thereby improving the accuracy, efficiency and robustness of information processing; and improves the quality and effect of decision-making by using the simulated annealing particle swarm algorithm for intelligent decision-making on energy consumption control.

[0101] Example 2 is an embodiment of the present invention, which provides an intelligent industrial energy consumption analysis and control decision-making method, including: obtaining energy consumption information, and transmitting and storing the energy consumption information; performing feature extraction and information conversion on the energy consumption information to obtain processed energy consumption information; performing analysis based on the energy consumption information and optimizing the analysis results; using the processed energy consumption information to establish a prediction model to predict energy consumption trends; evaluating the prediction results based on the energy consumption trends, and formulating energy consumption management strategies and decisions.

[0102] Example 3 is an embodiment of the present invention, which is different from the previous embodiment in that:

[0103] like Figure 2 As shown, if the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0104] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0105] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0106] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0107] Example 4 is an embodiment of the present invention, which provides an intelligent industrial energy consumption analysis and control decision-making system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0108] This example uses congestion control delay optimization algorithms, HDFS distributed storage technology, and deep learning network algorithm technology to achieve innovations and improvements in energy consumption information transmission, storage, processing, and analysis, as well as energy consumption forecasting and control. It then builds an intelligent industrial energy consumption analysis and control decision-making system to address the problems of slow information transmission and storage, information loss, low information processing efficiency, and low prediction accuracy in existing industrial energy consumption analysis and control decision-making systems. It also further improves the accuracy and stability of energy consumption information transmission and processing efficiency, providing technical support for the continuous improvement of industrial energy consumption control.

[0109] In terms of information transmission, this example uses a congestion control delay optimization algorithm for information transmission and compares it with the traditional method. The specific data is shown in Table 1.

[0110] Table 1 Comparison of information transmission of the present invention and traditional methods

[0111] Comparison Dimensions How to apply congestion control algorithms Commonly used transmission methods in the past Packet loss rate 3.4％ 7.6％ Maximum delay 1-3ms 10-20ms Resource allocation efficiency 96.1％ 91.7％ Transfer rate 15Gbps 7Gbps

[0112] As can be seen from Table 1, the present invention applies a congestion control algorithm for information transmission. Compared with the commonly used transmission methods in the past, it has the advantages of lower packet loss rate, shorter latency, higher resource allocation efficiency and faster transmission rate, thereby significantly improving the information transmission efficiency and achieving efficient and stable information transmission.

[0113] In terms of information processing, Table 2 shows a comparison between the DPTCN model used in the present invention and the existing model.

[0114] Table 2 Comparison of information processing effects of different models

[0115]

[0116] As can be seen from Table 2, the present invention uses the DPTCN model for information processing. Compared with other models, the DPTCN model has faster processing speed, better classification accuracy, stronger judgment, and better sorting ability.

[0117] In terms of energy consumption trend prediction, Table 3 shows comparative data between the prediction model used in the present invention and the traditional prediction model under various data sets.

[0118] Table 3 Comparison of decision error values ​​between traditional model and the proposed model

[0119] Dataset Traditional forecasting models Energy consumption prediction model of the present invention Data set 1 8.76％ 2.46％ Data set 2 7.84％ 3.67％ Data set 3 12.36％ 2.84％ average value 9.65％ 2.99％

[0120] As can be seen from Table 3, the decision error value of the energy consumption prediction model of the present invention is significantly lower than that of the traditional model, indicating that the present invention can make a more effective and stable decision-making solution.

[0121] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent industrial energy consumption analysis and control decision-making system, characterized by: include: Energy consumption information transmission and storage module, energy consumption information processing and analysis module, and energy consumption control and decision-making module; The energy consumption information transmission and storage module includes an information transmission unit for transmitting energy consumption information; an information storage unit for storing energy consumption information; The energy consumption information processing and analysis module includes an information processing unit for performing feature extraction and information conversion on energy consumption information; An information analysis unit, used to analyze energy consumption information and optimize analysis results; The energy consumption control and decision module includes an energy consumption prediction unit for establishing a prediction model using energy consumption information to predict energy consumption trends; The energy consumption control unit is used to formulate energy consumption management strategies and make decisions based on energy consumption trends.

2. The intelligent industrial energy consumption analysis and control decision-making system according to claim 1, characterized in that: The transmission of energy consumption information refers to the efficient transmission of information using the congestion control algorithm in the delay optimization algorithm; Using congestion control algorithms for information transmission includes determining information transmission path parameters, calculating delay-tolerant network bandwidth, calculating information transmission rate, and optimizing congestion status.

3. The intelligent industrial energy consumption analysis and control decision-making system according to claim 2, characterized in that: The storage of energy consumption information refers to the use of HDFS distributed storage technology to efficiently store information, including file segmentation operations, meta-information management operations, and information access and operations.

4. The intelligent industrial energy consumption analysis and control decision-making system according to claim 3, characterized in that: The feature extraction and information conversion of energy consumption information refers to the use of the DPTCN model in the deep learning network to perform key feature extraction and information conversion in the information processing task; The DPTCN model consists of a conversion layer, a squeeze layer, an excitation layer, and a recalibration layer.

5. The intelligent industrial energy consumption analysis and control decision-making system according to claim 4, characterized in that: Analyzing energy consumption information and optimizing analysis results refers to using information mining technology to identify information patterns and regularities, and using reinforcement learning to optimize analysis results; The use of information mining technology to identify information patterns and regularities refers to using information mining technology to discover hidden patterns and regularities from energy consumption information, provide decision-making basis for energy consumption management and operation, and apply association rules to mine the correlation between energy consumption and industry; The method of optimizing the analysis results using reinforcement learning refers to defining the state space and action space based on the actual energy consumption information, including the following steps: Use reinforcement learning algorithms for training and optimization based on the defined environment model, state space, action space, and reward function; Through interaction with the environment, the reinforcement learning algorithm continuously updates the strategy and adjusts the action selection strategy; By comparing with the real environment or through simulation, the reliability of the analysis results can be evaluated, and the analysis results can be adjusted and improved.

6. The intelligent industrial energy consumption analysis and control decision-making system according to claim 5, characterized in that: The use of energy consumption information to establish a prediction model refers to the energy consumption prediction unit using the processed energy consumption information to establish a prediction model to predict future energy consumption trends, so that regulators can make corresponding adjustments and decisions; The energy consumption trend prediction includes the following steps: Extract the features required for energy consumption prediction based on actual conditions; Use deep learning framework to build LSTM model; The LSTM model is trained using the training set. During the training process, the model parameters are optimized through the back propagation algorithm and loss function. The performance of the trained LSTM model is evaluated using the test set, and the error index between the predicted results and the actual energy consumption values ​​is calculated; Predict future energy consumption based on the trained LSTM model, convert the features into an input format acceptable to the model, and use the model to generate the corresponding energy consumption prediction results; The LSTM model is optimized and adjusted based on the prediction results and performance evaluation feedback.

7. The intelligent industrial energy consumption analysis and control decision-making system according to claim 6, characterized in that: Formulating energy consumption management strategies and decisions based on energy consumption trends refers to making intelligent energy consumption management decisions based on simulated annealing particle swarm algorithm; The energy consumption control intelligent decision-making based on simulated annealing particle swarm algorithm includes the following steps: The optimal solution of the PSO algorithm is used as the initial value and the initial annealing temperature is set. The initial position and speed of the factor are evaluated and compared according to the objective function. Calculate the fitness function difference between the new solution and the old solution, and determine whether to accept the new solution according to the Metropolis judgment standard; Calculate annealing temperature; Get a new energy consumption plan based on the updated position; The idea of ​​simulated annealing algorithm is introduced based on energy consumption decision-making, which avoids falling into local optimal solution by accepting differential solution with a certain probability. The algorithm is terminated according to the set termination conditions and the optimal energy consumption plan and the corresponding energy consumption value are output.

8. A method for analyzing and controlling energy consumption in intelligent industries, based on the system for analyzing and controlling energy consumption in intelligent industries according to any one of claims 1 to 7, characterized in that: include, Acquire energy consumption information, and transmit and store the energy consumption information; Perform feature extraction and information conversion on energy consumption information to obtain processed energy consumption information; Analyze and optimize the analysis results based on energy consumption information; Use the processed energy consumption information to establish a prediction model to predict energy consumption trends; Evaluate energy consumption trends based on forecast results and formulate energy consumption management strategies and decisions.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent industrial energy consumption analysis and control decision-making system according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent industrial energy consumption analysis and control decision-making system according to any one of claims 1 to 7 are implemented.

Images