AI-driven industrial decarburization process control method and system based on feedback regulation

Abstract

The invention provides an AI-driven industrial decarburization process control method and system based on feedback regulation, and relates to the technical field of industrial control, and the method comprises the steps: executing industrial decarburization process monitoring, and obtaining a process operation data stream sequence; performing process working condition time sequence iterative analysis; matching is carried out in a pre-constructed causal knowledge base; traversing the associated causal clue set to perform historical data backtracking exception investigation to obtain an exceptional associated causal clue set and an exception degree set; and performing feedback control scheme identification to obtain a target feedback control scheme, generating a feedback adjustment instruction, and issuing the feedback adjustment instruction to the control execution module to execute the target feedback control scheme. According to the method and the device, the technical problem that in the prior art, due to the fact that static model prediction or empirical rule-based judgment is mainly adopted, real-time tracing and dynamic intervention of process-level carbon emission abnormity are difficult to achieve, and the carbon emission control effect in a complex environment is not ideal is solved, and the control effect of the industrial decarburization process is improved.

Description

Technical Field

[0001] The present application relates to the field of industrial control technology, and in particular to an AI-driven industrial decarbonization process control method and system based on feedback regulation. Background Art

[0002] With the continuous development of industrial automation and environmental protection technologies, the requirements for controlling energy consumption and carbon emissions in industrial production processes are increasing. However, existing industrial decarbonization process control methods mainly rely on static model predictions or judgments based on empirical rules, which have obvious shortcomings in dealing with changes in carbon emissions under complex working conditions. The static model prediction method cannot fully reflect the dynamic changes in the production process, resulting in low prediction accuracy. In the actual production process, factors such as equipment parameters, energy consumption data, and environmental disturbance factors often change continuously, resulting in a large deviation between the prediction results of the static model and the actual situation. The judgment method based on empirical rules relies on the experience of the operator and is difficult to standardize and scale. When faced with complex working conditions, due to the lack of systematic modeling and analysis tools, it is difficult to find the precise root cause when carbon emissions are abnormal, resulting in unsatisfactory carbon emission control effects.

[0003] In summary, the existing technology has the following technical problems: it mainly relies on static model prediction or judgment based on empirical rules, lacks the ability to systematically model the causes and manifestations of carbon emission changes under complex working conditions, and is difficult to achieve real-time tracing and dynamic intervention of process-level carbon emission anomalies, resulting in unsatisfactory carbon emission control effects in complex environments. Summary of the Invention

[0004] The purpose of this application is to provide an AI-driven industrial decarbonization process control method and system based on feedback regulation, so as to solve the technical problems in the existing technology that the existing technology mainly relies on static model prediction or judgment based on empirical rules, lacks the ability to systematically model the causes and manifestations of carbon emission changes under complex working conditions, and is difficult to achieve real-time tracing and dynamic intervention of process-level carbon emission anomalies, resulting in unsatisfactory carbon emission control effects in complex environments.

[0005] In view of the above problems, this application provides an AI-driven industrial decarbonization process control method and system based on feedback regulation.

[0006] In the first aspect, the present application provides an AI-driven industrial decarbonization process control method based on feedback regulation, which is implemented by an AI-driven industrial decarbonization process control system based on feedback regulation, wherein the AI-driven industrial decarbonization process control method based on feedback regulation includes: executing industrial decarbonization process monitoring to obtain a process operation data flow sequence, wherein each process operation data flow includes equipment parameters, energy consumption data, carbon emission values ​​and environmental disturbance factors; traversing the process operation data flow sequence to perform iterative analysis of process operating conditions to obtain iterative abnormal operating condition state characteristics; matching the iterative abnormal operating condition state characteristics in a pre-built causal knowledge base to obtain a set of associated causal clues; traversing the associated causal clue set to perform historical data backtracing anomaly investigation to obtain an abnormal associated causal clue set and an abnormality degree set; identifying feedback control schemes based on the abnormal associated causal clue set and the abnormality degree set to obtain a target feedback control scheme, and generating feedback adjustment instructions, and sending the feedback adjustment instructions to the control execution module to execute the target feedback control scheme.

[0007] Optionally, a multimodal sensing array is deployed at key nodes of the target industrial decarbonization process, wherein the multimodal sensing array includes an equipment parameter collector, an energy consumption metering unit, a carbon emission monitoring device and an environmental disturbance sensing module; the multimodal sensing array is traversed to perform simultaneous sequential perception of key nodes to construct a time-synchronized process operation data flow sequence.

[0008] Optionally, a first process operation data flow is extracted from the process operation data flow sequence; the operating condition state of the first process operation data flow is identified based on a preset process operating condition standard value to obtain a first process operation operating condition state vector, and the preset process operating condition standard value and the first process operation operating condition state vector are added to a first iterative memory unit; based on the first iterative memory unit, a process operating condition timing iterative analysis is performed on the second process operation data flow extracted from the process operation data flow sequence to obtain a second process operating condition state vector; based on the second process operating condition state vector, the first iterative memory unit is updated to obtain a second iterative memory unit, and so on, a timing iterative analysis is performed on the remaining process operation condition state vectors in the process operation data flow sequence to obtain a target iterative memory unit; abnormal feature identification is performed on the target iterative memory unit to obtain the iterative abnormal operating condition state feature.

[0009] Optionally, a preset process operating condition standard value is extracted from the first iterative memory unit, and the operating condition state is identified on the second process operation data stream to obtain a second initial process operation condition state vector; the first process operation condition state vector is extracted from the first iterative memory unit, and the second initial process operation condition state vector is time-series iteratively enhanced to obtain a second process operation condition state vector.

[0010] Optionally, the element similarity between the first process operating condition state vector and the second initial process operating condition state vector is calculated to obtain an element similarity set; matrix processing is performed based on the element similarity set to construct a timing iteration enhancement matrix; and convolution enhancement is performed on the second initial process operating condition state vector using the timing iteration enhancement matrix to obtain the second process operating condition state vector.

[0011] Optionally, a set of historical abnormal related events of the industrial decarbonization process is obtained; the set of historical abnormal related events is traversed to extract causal clues, and a set of historical abnormal causal clues and a corresponding set of historical abnormal operating condition state features are obtained; the set of historical abnormal causal clues and the corresponding set of historical abnormal operating condition state features are integrated to obtain an integrated set of historical abnormal causal clues and a corresponding integrated set of historical abnormal operating condition state features; the integrated set of historical abnormal causal clues and the corresponding integrated set of historical abnormal operating condition state features are associated and stored to obtain the causal knowledge base.

[0012] Optionally, the historical abnormal operating condition state feature set is integrated in the same category, and the integration results are respectively averaged to obtain an integrated historical abnormal operating condition state feature set; based on the integrated historical abnormal operating condition state feature set, the historical abnormal causal clue set is mapped and integrated, and the mapping integration results are unioned to obtain the integrated historical abnormal causal clue set.

[0013] Optionally, in the historical data backtracking window, the associated causal clue set is traversed to perform historical data backtracking to obtain an associated causal clue historical data sequence set; the associated causal clue historical data sequence set is traversed to perform standard value deviation identification to obtain an associated causal clue deviation value sequence set; the associated causal clue set is checked for abnormalities based on the associated causal clue deviation value sequence set to obtain an abnormal associated causal clue set; the abnormal associated causal clue deviation value sequence corresponding to the abnormal associated causal clue set is extracted from the associated causal clue deviation value sequence set to perform deviation value mean processing to obtain the abnormality degree set.

[0014] Optionally, a feedback control scheme identifier is pre-built, and the feedback control scheme identifier is used to identify the abnormal correlation causal clue set and the abnormality degree set to obtain the target feedback control scheme.

[0015] In the second aspect, the present application also provides an AI-driven industrial decarbonization process control system based on feedback regulation, which is used to execute the AI-driven industrial decarbonization process control method based on feedback regulation as described in the first aspect, wherein the AI-driven industrial decarbonization process control system based on feedback regulation includes: a process monitoring module, which is used to perform industrial decarbonization process monitoring and obtain a process operation data flow sequence, wherein each process operation data flow includes equipment parameters, energy consumption data, carbon emission values ​​and environmental disturbance factors; a timing iteration analysis module, which is used to traverse the process operation data flow sequence to perform process operating condition timing iteration analysis and obtain iterative abnormal operating condition state characteristics; a knowledge base matching module, which is used to match the iterative abnormal operating condition state characteristics in a pre-built causal knowledge base to obtain a set of associated causal clues; an anomaly investigation module, which is used to traverse the associated causal clue set to perform historical data backtracing anomaly investigation and obtain an abnormal associated causal clue set and an abnormality degree set; a control scheme execution module, which is used to identify feedback control schemes based on the abnormal associated causal clue set and the anomaly degree set, obtain a target feedback control scheme, and generate feedback adjustment instructions, and send the feedback adjustment instructions to the control execution module to execute the target feedback control scheme.

[0016] One or more technical solutions provided in this application have at least the following beneficial effects: By executing industrial decarbonization process monitoring, a process operation data flow sequence is obtained, wherein each process operation data flow includes equipment parameters, energy consumption data, carbon emission values ​​and environmental disturbance factors; the process operation data flow sequence is traversed to perform iterative analysis of process operating conditions, and the iterative abnormal operating condition state characteristics are obtained; based on the iterative abnormal operating condition state characteristics, a match is performed in a pre-built causal knowledge base to obtain a set of associated causal clues; the associated causal clue set is traversed to perform historical data backtracing anomaly investigation to obtain an abnormal associated causal clue set and an abnormality degree set; feedback control scheme identification is performed based on the abnormal associated causal clue set and the abnormality degree set to obtain a target feedback control scheme, and a feedback adjustment instruction is generated, and the feedback adjustment instruction is sent to the control execution module to execute the target feedback control scheme. In other words, by collecting the process operation data flow sequence, conducting iterative analysis of the process operating conditions, extracting the state characteristics of abnormal operation, and using the pre-built causal knowledge base to trace the causes of carbon emission anomalies, and through historical data backtracking, the causes of the anomalies are associated with the actual operating conditions, and the target feedback control scheme is determined to dynamically intervene in the decarbonization process, achieving accurate tracing and efficient control of carbon emission anomalies, and greatly improving the control effect of the industrial decarbonization process.

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0019] Figure 1 This is a flow chart of the AI-driven industrial decarbonization process control method based on feedback regulation in this application.

[0020] Figure 2 This is a structural diagram of the AI-driven industrial decarbonization process control system based on feedback regulation in this application.

[0021] Explanation of the accompanying symbols: process monitoring module 11, time sequence iteration analysis module 12, knowledge base matching module 13, abnormality troubleshooting module 14, control scheme execution module 15. DETAILED DESCRIPTION

[0022] This application provides an AI-driven industrial decarbonization process control method and system based on feedback regulation, which solves the technical problem in the prior art that the existing technology mainly relies on static model prediction or judgment based on empirical rules, lacks the ability to systematically model the causes and manifestations of carbon emission changes under complex working conditions, and is difficult to achieve real-time tracing and dynamic intervention of process-level carbon emission anomalies, resulting in unsatisfactory carbon emission control effects in complex environments. By collecting process operation data flow sequences, performing iterative analysis of process operating conditions, extracting state characteristics of abnormal operations, and using a pre-built causal knowledge base to trace the causes of carbon emission anomalies, and by tracing back historical data, correlating the causes of anomalies with actual working conditions, determining the target feedback control scheme to dynamically intervene in the decarbonization process, accurate tracing and efficient control of carbon emission anomalies are achieved, greatly improving the control effect of the industrial decarbonization process.

[0023] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0024] For example, see the attached Figure 1 The present application provides an AI-driven industrial decarbonization process control method based on feedback regulation, wherein the AI-driven industrial decarbonization process control method based on feedback regulation is executed by an AI-driven industrial decarbonization process control system based on feedback regulation, and the AI-driven industrial decarbonization process control method based on feedback regulation specifically includes the following steps: S100: Execute industrial decarbonization process monitoring to obtain a process operation data flow sequence, wherein each process operation data flow includes equipment parameters, energy consumption data, carbon emission values, and environmental disturbance factors.

[0025] Furthermore, the present application S100 includes: A multimodal sensing array is deployed at the key nodes of the target industrial decarbonization process, wherein the multimodal sensing array includes an equipment parameter collector, an energy consumption metering unit, a carbon emission monitoring device and an environmental disturbance perception module; the multimodal sensing array is traversed to perform simultaneous sequential perception of the key nodes to construct a time-synchronized process operation data flow sequence.

[0026] Specifically, multimodal sensing arrays are deployed at key nodes in the industrial decarbonization process to comprehensively monitor various factors that may affect carbon emissions during the production process. The multimodal sensing array includes equipment parameter collectors, energy consumption metering units, carbon emission monitoring equipment, and environmental disturbance sensing modules. Each sensing device is responsible for collecting different types of data in real time. For example, the equipment operation parameter collector is used to collect temperature, pressure, flow rate, speed, valve opening, etc.; the energy consumption metering unit is used to connect to current transformers, calorimeters, gas flow meters, etc. to obtain real-time energy consumption data; the carbon emission monitoring equipment is used to collect carbon emission values ​​using CO2 concentration detectors, flue gas analyzers, and infrared sensor modules; environmental disturbance sensing modules include temperature and humidity sensors, raw material batch recorders, and external power supply stability detectors.

[0027] Through a multimodal sensing array, each key node in the target industrial decarbonization process is monitored simultaneously across multiple dimensions, a process known as simultaneous sensing. All sensing devices at each key point operate simultaneously, collecting data and ensuring that this data is synchronized in time. This multimodal sensing array generates a time-synchronized process operation data stream sequence. This integrates collected information such as equipment parameters, energy consumption data, carbon emissions, and environmental disturbance factors into a unified process operation data stream sequence in chronological order, encompassing time-synchronized data collected from each sensing device. Time synchronization means that data from different sources is arranged in the same time unit, ensuring data order and temporal consistency and avoiding analytical errors caused by time deviations.

[0028] In the industrial decarbonization process, energy issues must be addressed first, as energy production and consumption are the core source of carbon emissions. Sustainable energy solutions can be implemented by replacing traditional fossil fuels with clean energy sources such as photovoltaics, hydrogen, and wind power, effectively reducing carbon emissions from energy consumption and direct carbon emissions by 30%-60%. For example, a steel company implemented a wind-solar hybrid power generation system, reducing annual coal consumption by 120,000 tons and lowering carbon emissions intensity by 42%. A multimodal sensing array monitors energy consumption and carbon emissions in real time. Energy metering units monitor clean energy usage, while carbon emission monitoring equipment records carbon emissions from the conversion from traditional to clean energy sources. Furthermore, data from waste heat recovery systems, combined with microgrid energy distribution, ensures efficient energy utilization and minimizes carbon emissions. For example, a steel company reduced energy consumption by 30% and, in turn, carbon emissions by 42% through waste heat recovery technology. By deploying a multimodal sensing array, comprehensive, real-time monitoring of all factors affecting carbon emissions ensures complete and accurate monitoring data.

[0029] S200: Traversing the process operation data flow sequence to perform iterative analysis of the process working condition time sequence to obtain iterative abnormal operating condition state characteristics.

[0030] Furthermore, the present application S200 includes: Extract a first process operation data flow from the process operation data flow sequence; perform operating condition identification on the first process operation data flow based on a preset process operating condition standard value to obtain a first process operation operating condition state vector, and add the preset process operating condition standard value and the first process operation operating condition state vector into a first iterative memory unit; perform process operating condition time series iterative analysis on the second process operation data flow extracted from the process operation data flow sequence based on the first iterative memory unit to obtain a second process operating condition state vector; update the first iterative memory unit based on the second process operating condition state vector to obtain a second iterative memory unit, and so on, perform time series iterative analysis on the remaining process operation condition state vectors in the process operation data flow sequence to obtain a target iterative memory unit; perform abnormal feature identification on the target iterative memory unit to obtain the iterative abnormal operating condition state feature.

[0031] Specifically, the process operation data stream sequence is traversed, and the first process operation data stream is randomly extracted. This data stream contains information such as equipment parameters, energy consumption data, and carbon emissions for a specific time period. Based on the industrial equipment, process flow, and actual operating conditions, a preset process operating condition standard value is set, representing the parameter range that the equipment and process should achieve under normal operating conditions. Using this preset process operating condition standard value, the operating condition of each parameter in the first process operation data stream is identified. By comparing the actual value in the first process operation data stream with the preset process operating condition standard value, the current equipment and process are determined to be normal. For example, if the standard value indicates that the temperature of a piece of equipment in a chemical plant is between 700°C and 950°C, and the actual temperature is 850°C, then the equipment is in a normal state; otherwise, it may be in an abnormal state. The first process operation condition state vector is a mathematical vector that describes the first process operation data stream and contains the values ​​of various equipment operating parameters (such as temperature, pressure, and flow rate). Each state vector represents the operating state at a point in time.

[0032] The first process operating condition state vector is stored in the first iterative memory unit along with the preset process condition standard value. By continuously iteratively analyzing new data and comparing the operating condition state vector with the standard value, the data in the memory unit is gradually optimized. The current state may be caused by other equipment or other processes, so other clues need to be investigated for impact. If any problems exist, feedback adjustments are required.

[0033] A second process operation data stream is extracted from the process operation data stream sequence. This is the process operation data stream at the next moment after the first process operation data stream. Based on the first iterative memory unit, an iterative process operating condition time series analysis is performed on the second process operation data stream. Specifically, the operating condition state vector of the first process is combined with the second process operation data stream to determine the operating condition of the second process and optimize it.

[0034] The first iterative memory unit is updated by the second process operating condition state vector to combine the operating condition information of the second process with the operating condition information of the first process, so that the understanding of the entire process is more comprehensive. The second process operating condition state vector and the preset process operating condition standard value are added to the first iterative memory unit to form a second iterative memory unit. For example, if the first iterative memory unit stores the first process operating condition state vector and the preset process operating condition standard value, then during the update process, the second process operating condition state vector and the preset process operating condition standard value will also be stored in the second iterative memory unit. The second iterative memory unit is a new memory unit obtained based on the update of the first iterative memory unit. Each time new process operating data (such as the second process operating condition state vector) is added, the first iterative memory unit will be updated to form a new iterative memory unit.

[0035] In this way, we continue to traverse the process operation data flow sequence, extract subsequent data, and gradually perform iterative analysis of the process operating conditions in a timely manner to include richer data. Each iteration generates a new iterative memory unit based on the previous iterative memory unit and the current process operation condition state vector. For example, if there are 100 data flows in the process operation data flow sequence, 99 iterations will be performed to eventually obtain the target iterative memory unit. During each iteration, the information in the previous iterative memory unit will be used to perform time-series iterative enhancement on the current process operation condition state vector to obtain more accurate condition state information. Time-series iterative analysis is a method of analyzing current data based on historical data, using information from the previous time point to predict and optimize future operating conditions. As the data flow continues to iterate and update, all process operation data are analyzed cyclically until the final target iterative memory unit is obtained.

[0036] The target iterative memory unit contains all the data that has undergone time-series iterative analysis. It is a comprehensive representation of the entire process and includes the time-series evolution and change trends of all key operating condition information (such as temperature, pressure, flow, etc.) of the process system. By conducting in-depth analysis of the historical operating condition data in the iterative memory unit, characteristics that do not conform to the normal operating mode or potential abnormal trends are identified. Abnormal characteristics are usually manifested as drastic changes in certain operating condition values ​​or exceeding the set standard range, indicating problems such as equipment failure, process instability, or low energy efficiency. It is usually determined whether the data exceeds the normal tolerance range by using the deviation from the standard deviation and the mean. Abnormal characteristics usually include those operating conditions that deviate from expectations, fail, or are abnormal.

[0037] When the abnormal feature identification is completed, the abnormal operating condition features are extracted according to the identified abnormal state to obtain the iterative abnormal operating condition state features, which describe the deviation of certain operating conditions from the normal state during the iteration process, including certain equipment failures, system abnormalities or abnormal operating behaviors, revealing abnormal or atypical operating conditions in the industrial process.

[0038] Abnormal feature recognition, as a dynamic feedback mechanism, is a key component of process optimization. By detecting anomalies in real-time data during the production process, potential problems in the process can be promptly identified, especially those not detected by traditional process parameter adjustments. Process optimization is the reprogramming of the heart of industry, and process reengineering unlocks hidden carbon reduction potential. Traditional industrial processes are plagued by significant energy waste and redundant links. Through process parameter optimization, lean production management, and supply chain collaboration, energy reduction can be achieved throughout the entire process chain. For example, an automobile manufacturer used the process diagnostic tools in the SPARK theory to identify 20% energy waste in the casting process. After the transformation, energy consumption per unit product was reduced by 18%.

[0039] The SPARK concept is a theoretical framework specifically designed for carbon reduction in the industrial sector. It addresses the challenges faced by industry (process industries, which account for 72% of industrial carbon emissions) in achieving decarbonization goals. The SPARK concept deeply integrates five key elements: sustainable energy, process optimization, AI analysis, real-time control, and knowledge systems. This framework forms a co-evolutionary framework, providing a systematic solution for industrial carbon reduction. The five core elements of SPARK include S, P, A, R, and K. Among them, S stands for sustainable energy, that is, replacing traditional fossil energy with clean energy such as photovoltaics and hydrogen energy, combined with waste heat recovery, microgrid deployment and other technologies, which can reduce direct carbon emissions by 30%-60%; P stands for process optimization, which can achieve full-link energy conservation through process parameter optimization, lean production management and supply chain collaboration; A stands for AI analysis, which uses machine vision, deep learning and other technologies to conduct real-time monitoring, prediction and optimization scheduling. For example, the real-time energy consumption monitoring system based on machine vision can dynamically adjust the operating status of equipment, combine deep learning to predict carbon emission peaks, and optimize production scheduling; R stands for real-time control, which realizes closed-loop response, dynamically adjusts equipment parameters (such as temperature and pressure), and quickly controls carbon emissions; K stands for knowledge embedding, which is used to standardize the packaging of modules such as energy management, process parameters, and AI models to form a cross-industry knowledge graph.

[0040] The process of iteratively analyzing process conditions in a time series of process data streams is equivalent to the P factor in SPARK theory. Real-time monitoring and analysis of process conditions can identify potential issues within the process and enable timely adjustments, ensuring production stability and avoiding disruptions or safety incidents caused by unexpected failures. By continuously optimizing and updating the target iterative memory unit, conditions with high energy consumption or carbon emissions are identified, and adjustments are made based on these abnormal data, achieving energy efficiency improvements and emissions reductions throughout the entire process.

[0041] Furthermore, the present application further comprises the following steps: Extract the preset process operating condition standard value from the first iterative memory unit, perform operating condition identification on the second process operation data stream, and obtain a second initial process operation condition state vector; extract the first process operation condition state vector from the first iterative memory unit, perform time series iterative enhancement on the second initial process operation condition state vector, and obtain a second process operation condition state vector.

[0042] Calculate the element similarity between the first process operating condition state vector and the second initial process operating condition state vector to obtain an element similarity set; perform matrix processing based on the element similarity set to construct a timing iteration enhancement matrix; use the timing iteration enhancement matrix to perform convolution enhancement on the second initial process operating condition state vector to obtain the second process operating condition state vector.

[0043] Specifically, the preset process operating condition standard values ​​are extracted from the first-iteration memory unit, representing the normal ranges of various key parameters in the production process. Based on these preset process operating condition standard values, the operating condition of the second process operation data stream is identified. The process status is determined by comparing the actual parameters in the second process operation data stream with the preset operating condition standard values. For example, if the temperature value exceeds the preset standard value, the operating condition may be identified as an overload; if the flow rate exceeds the set range, it may be an abnormal state. The core of operating condition identification is to identify the deviation between each process parameter (such as temperature and pressure) and the standard value. After operating condition identification, a preliminary state vector containing the second process operation data stream is obtained. For example, the temperature is 850°C, the pressure is 28kPa, the flow rate is 55L / min, and the carbon dioxide concentration is 13.1%.

[0044] The first process operating condition state vector is extracted from the first iterative memory unit, and the second initial process operating condition state vector is subjected to time-series iterative enhancement. That is, based on the time relationship between the first process operating data stream and the second process operating data stream, the states of the two are compared, optimized, and enhanced to identify possible deviations or anomalies in the second process operating data stream, while improving the accuracy and robustness of the state vector. Time-series iterative enhancement refers to a dynamic adjustment process that takes into account the relationship between the previous and next data streams and improves the accuracy of operating condition identification through iterative updates. After time-series iterative enhancement, the result is the second process operating condition state vector, which is more accurate than the second initial process operating condition state vector and can reflect a more realistic operating condition.

[0045] Specifically, the element similarity between the first process operating condition state vector and the second initial process operating condition state vector. Each operating condition state vector contains multiple elements (such as temperature, pressure, etc.), so it is necessary to calculate the similarity between each pair of corresponding elements. Element similarity is used to quantify the similarity between the corresponding elements in the first process operating condition state vector and the second initial process operating condition state vector, helping to identify the similarities and differences between the two operating condition state vectors. For example, the similarity of temperature values, the similarity of pressure values, etc. The distance between two vectors in space is measured by Euclidean distance. A smaller Euclidean distance indicates that the two vectors are more similar. According to the different elements in the operating condition state vector, the most suitable similarity calculation method is selected. For each pair of parameters (such as temperature, pressure, flow rate, etc.), the calculated similarity value is stored in the element similarity set. For example, the state vector for the first process's operating conditions is: temperature 880°C, pressure 40 kPa, flow rate 60 L / min, and CO2 concentration 12%; the state vector for the second initial process's operating conditions is: temperature 890°C, pressure 40.1 kPa, flow rate 61 L / min, and CO2 concentration 12.2%. Calculating element similarity yields a temperature similarity of 10, a pressure similarity of 0.1, a flow rate similarity of 1, and a CO2 concentration similarity of 0.2.

[0046] Normalize the element similarity set, converting the similarity values ​​to a standard range (e.g., [0, 1]). Then, populate the processed data into an initially empty matrix to generate a time series iterative enhancement matrix. Each row and column in the matrix represents the process state in a different time period, and each element in the matrix corresponds to the similarity between two operating conditions. The time series iterative enhancement matrix is ​​obtained through matrixization, forming a matrix that can describe the changing patterns of time series data. This matrix is ​​used to enhance the characteristics of time series data and further improve data accuracy and reliability.

[0047] The second initial process operating condition state vector is convolved using a time-series iteration enhancement matrix. Convolution is used to extract local features and temporal patterns from data. By convolving the convolution kernel with the time-series matrix, the temporal variation patterns in the data are enhanced. Convolution enhancement is a signal processing technique used to enhance local features in data. When processing the operating condition state vector, convolution enhancement strengthens the temporal characteristics of the data by convolving the time-series iteration enhancement matrix, improving the accuracy and robustness of the prediction. Convolution typically uses a sliding window approach, performing a dot product between the convolution kernel (a small matrix) and the input matrix and moving the window to capture temporal characteristics. Convolution enhancement can correlate operating conditions at adjacent moments, enhancing their temporal information. After convolution enhancement, the optimized second process operating condition state vector is obtained, which reflects the results of the improvements and optimization of the operating conditions in the second time period. For example, after convolution enhancement, the state vector of the second initial process operating condition (temperature is 890°C, pressure is 40.1kPa, flow rate is 61L / min, and CO2 concentration is 12.2%) finally obtains temperature of 889°C, pressure of 40.5kPa, flow rate of 61.2L / min, and CO2 concentration of 12.5%.

[0048] By extracting the preset operating condition standard value from the first iterative memory unit and combining it with the time series iteration enhancement method, the operating condition state vector of the current process can be accurately identified and adjusted, making the judgment of the current operating condition more accurate and providing strong data support for subsequent optimization, thereby improving the efficiency and stability of the entire industrial decarbonization and optimization process.

[0049] S300: Matching is performed in a pre-built causal knowledge base based on the iterative abnormal operating condition state characteristics to obtain a set of associated causal clues.

[0050] Furthermore, the present application S300 includes: Obtain a set of historical abnormal related events of the industrial decarbonization process; traverse the set of historical abnormal related events to extract causal clues, and obtain a set of historical abnormal causal clues and a corresponding set of historical abnormal operating condition state features; integrate the set of historical abnormal causal clues and the corresponding set of historical abnormal operating condition state features to obtain an integrated set of historical abnormal causal clues and a corresponding integrated set of historical abnormal operating condition state features; associate and store the integrated set of historical abnormal causal clues and the corresponding integrated set of historical abnormal operating condition state features to obtain the causal knowledge base.

[0051] Furthermore, the present application further comprises the following steps: The historical abnormal operating condition state feature set is integrated in the same category, and the integration results are respectively averaged to obtain the integrated historical abnormal operating condition state feature set; based on the integrated historical abnormal operating condition state feature set, the historical abnormal causal clue set is mapped and integrated, and the mapping integration results are unioned to obtain the integrated historical abnormal causal clue set.

[0052] Specifically, through equipment self-check systems, operator reports, and automated monitoring systems, we capture all past abnormal events related to industrial decarbonization processes. Each event includes the operating conditions at the time of the abnormality and possible causal relationships, resulting in a collection of historical abnormal correlation events. By traversing this collection of historical abnormal correlation events, we extract causal clues and identify the causal relationships between abnormal events. For example, if a piece of equipment's pressure exceeds the specified limit, leading to a temperature increase, this is a typical causal relationship.

[0053] By analyzing the interrelationships between abnormal events, the root causes behind the abnormalities can be identified. Using algorithms such as Apriori or FP-growth, it is possible to identify which abnormal events often occur together from the data. For example, it may be found that an overheated reactor temperature is always accompanied by a cooling system failure or a heating system overload. In the process of causal clue extraction, in addition to discovering the causal relationship between abnormal events, it is also necessary to extract the operating parameters (temperature, pressure, flow rate, etc.) at the time of each event. These parameters provide the context for the occurrence of the abnormal event. For example, in some abnormal events, excessive temperature, equipment overload, and equipment damage may be mentioned, and the corresponding operating characteristics may be a reactor temperature of 750°C, a pressure of 15kPa, and a flow rate of 50L / min.

[0054] All causal relationships extracted from historical abnormal events are aggregated to form a set of historical abnormal causal clues. This not only shows the temporal sequence between certain abnormal events, but may also reveal the underlying causes. For each abnormal event, relevant operating condition data is collected, including parameters such as temperature, pressure, flow, and energy consumption. For example, for an abnormal event involving excessive temperature, operating condition data such as temperature, pressure, fuel flow, and oxygen content may be recorded. These data can help analyze the cause of the excessive temperature. The set of historical abnormal causal clues is a set of causal relationships between all identified abnormal events. Each causal clue represents a causal relationship between one or more pairs of events, revealing the cause of the abnormal event. The set of historical abnormal operating condition state feature characteristics refers to the set of operating condition features related to each abnormal event. These features include parameters such as temperature, pressure, flow, and energy consumption. They are used to analyze the operating environment and equipment status when the abnormal event occurred, thereby revealing possible abnormal causes.

[0055] The historical abnormal operating condition state feature set is integrated in the same category, and the operating condition data with similar properties or types are merged or classified. For example, all the operating condition data about temperature are merged into one large category, and the pressure data are merged into another large category. Through the same category integration, the redundant information in the operating condition data is reduced and the main features are extracted. For each type of integrated operating condition data, the mean is calculated, that is, the operating condition data of the same category is averaged. Mean processing can reduce the impact of a single extreme value on the overall analysis and make the data more stable. For example, for all recorded temperature values, the average of these temperature values ​​is calculated to obtain the mean representing the operating condition data of this category. Through mean processing, an integrated historical abnormal operating condition state feature set is obtained, which contains the average operating condition data of each category and can represent the typical operating conditions under different categories.

[0056] Based on the integrated historical abnormal operating condition feature set, the historical abnormal causal clue set is mapped and integrated. Elements in the two data sets are associated or fused through a mapping relationship (such as a numerical range mapping or functional relationship). The integrated historical abnormal operating condition feature set serves as the basis for mapping the historical abnormal causal clue set to align the operating condition features with the causal clues. The features in the integrated historical abnormal operating condition feature set are aligned and mapped with the causal relationships in the historical abnormal causal clue set. Combining the operating condition features with the abnormal causal relationships creates a more comprehensive information set. The purpose of mapping and integration is to further improve and optimize the knowledge base by combining features and causal relationships. For example, if certain abnormal conditions (such as overtemperature) are always accompanied by certain causal clues (such as equipment failure), these two data points are associated. The goal of mapping is to unify the operating condition features and causal clues into a single set, forming a new dataset.

[0057] The mapping integration results are unioned, that is, all the causal clues obtained are merged and the duplicate parts are removed to form the final integrated historical abnormal causal clue set, which represents the possible causal relationships and their corresponding operating condition characteristics in all historical abnormal events. The integrated historical abnormal causal clue set is a new causal clue set obtained based on mapping integration and union, which contains the causal relationships of historical abnormal events, and these causal relationships correspond to the processed operating condition characteristics. The final integrated historical abnormal causal clue set and its corresponding integrated historical abnormal operating condition state feature set are stored in the causal knowledge base. The causal knowledge base is a database containing all abnormal events and their causal relationships, including abnormal events occurring under different operating conditions, their causes, and corresponding operating condition characteristics.

[0058] For example, when the temperature is 950°C, the pressure is 45kPa, and the flow rate is 70L / min, equipment failure occurs, and the corresponding cause is high temperature; when the temperature is 920°C, the pressure is 20kPa, and the flow rate is 80L / min, low efficiency occurs, and the corresponding causes are high temperature and abnormal pressure; when the temperature is 900°C, the pressure is 30kPa, and the flow rate is 25L / min, high energy consumption occurs, and the corresponding causes are high temperature and low flow. Extracting causal relationships from the set of historical abnormal events shows that excessive temperature corresponds to equipment failure, excessive temperature corresponds to low efficiency, excessive temperature corresponds to high energy consumption, and excessive temperature combined with pressure fluctuations corresponds to equipment failure. By building a causal knowledge base, we accumulate a wealth of historical experience and lessons, improving our ability to predict future abnormal events. When similar operating conditions occur in the future, we can quickly identify potential abnormalities and take appropriate preventive measures.

[0059] According to the state characteristics of iterative abnormal operating conditions, semantic matching mechanisms (such as vector distance and label matching) are used to search in the causal knowledge base to determine the set of related causal clues. Vector distance refers to converting the state characteristics of abnormal operating conditions into vector form and evaluating their similarity with known conditions in the causal knowledge base by calculating the distance between vectors (such as cosine similarity, Euclidean distance, etc.). If the operating condition characteristics have clear label classifications, the label matching mechanism is used to compare them with the labels in the causal knowledge base to find the relevant causal relationship. Based on the extracted state characteristics of iterative abnormal operating conditions, the selected matching mechanism is used to search in the causal knowledge base to find relevant causal clues. For example, if the current state characteristic is too high temperature, all causal clues related to too high temperature can be found through vector distance matching or label matching mechanism.

[0060] Based on the matching results, causal clues associated with the current abnormal operating condition are extracted from the causal knowledge base to obtain a collection of related causal clues. This helps understand the possible causes of the anomaly and further assists in developing targeted treatment strategies. By introducing vector distance matching and label matching mechanisms, the historical causal clues most relevant to the current abnormal state are accurately identified, thereby improving the accuracy of anomaly detection.

[0061] S400: Traverse the associated causal clue set to perform historical data backtracking anomaly investigation to obtain an abnormal associated causal clue set and an abnormality degree set.

[0062] Furthermore, the present application S400 includes: In the historical data backtracking window, the associated causal clue set is traversed to perform historical data backtracking to obtain the associated causal clue historical data sequence set; the associated causal clue historical data sequence set is traversed to perform standard value deviation identification to obtain the associated causal clue deviation value sequence set; the associated causal clue set is checked for abnormalities based on the associated causal clue deviation value sequence set to obtain the abnormal associated causal clue set; the abnormal associated causal clue deviation value sequence corresponding to the abnormal associated causal clue set is extracted from the associated causal clue deviation value sequence set to perform deviation value mean processing to obtain the abnormality degree set.

[0063] Specifically, a historical data lookback window is determined, meaning a historical time period is selected within the time series and retrospective analysis is performed on the data within that time period. Retrospective analysis is performed on historical data based on the defined historical data lookback window. The purpose of retrospective analysis is to compare current anomaly causal clues with historical data to gain a deeper understanding of the possible root causes of the anomaly. Based on the previously identified causal relationships, the corresponding historical data series are reviewed and analyzed to identify past trends and potential issues. Each causal clue may correspond to a specific production process or operation, and the associated historical data series may include variable data such as temperature, pressure, and flow rate for this process. Based on the lookback window and associated causal clues, historical data is extracted and a set of associated causal clue historical data series is formed. This set contains multiple data items related to each causal clue, forming a time series. The associated causal clue historical data series set is extracted from the associated causal clue set and is associated with each causal clue. Each data series contains historical data that occurred within the lookback window, helping to analyze and identify the cause of the anomaly.

[0064] For the historical data series of causal clues, standard deviation identification is performed. The deviation between the actual operating conditions and the standard operating conditions is calculated. This is compared with the preset process condition standard value to obtain a series of causal clue deviation values. Each deviation value in the series of causal clue deviation values ​​is compared with a preset threshold to determine the number of deviation values ​​greater than or equal to the preset threshold. If this number exceeds the allowed number, it is considered abnormal and identified as an abnormal causal clue, thus obtaining a series of abnormal causal clues. The preset threshold is a predefined value used to determine whether the deviation value deviates from the normal range. The preset threshold is used to filter out data with obvious anomalies. The allowed number is an acceptable threshold that controls how many deviation values ​​are allowed to exceed the normal range. For example, if only one of 10 consecutive measurements exceeds the threshold, it may be acceptable; however, if the number of deviations exceeds the allowed number, it is considered an anomaly. For example, if the identified sequence of deviation values ​​for associated causal clues is as follows: 80, 70, 85, 95, and 90, with a preset threshold of 80 and a maximum of three deviations allowed, then because the number of deviations greater than 80 exceeds the preset threshold of three, this is considered an abnormal causal clue set. The deviation values ​​for these abnormal causal clues are averaged to obtain (80 + 85 + 95 + 90) / 4 = 87.5.

[0065] The abnormal causal clue deviation value sequence corresponding to the abnormal causal clue set is extracted from the set of associated causal clue deviation value sequences. The deviation values ​​of the abnormal causal clue deviation value sequence are averaged, and the abnormal degree set of these abnormal states is calculated, reflecting the severity of the abnormality. The abnormal degree is used to quantify the severity of abnormal behavior. It is usually reflected by averaging the abnormal deviation values ​​to reflect the overall impact of the abnormality. A larger mean value indicates a more severe abnormality. Through historical data backtracking, standard value deviation identification, and anomaly investigation, potential anomalies can be effectively identified to avoid missed detections. The deviation value average is used to quantify the severity of the anomaly and determine whether emergency measures or long-term optimization measures are needed.

[0066] S500: Identify feedback control schemes based on the abnormality-related causal clue set and the abnormality degree set, obtain a target feedback control scheme, generate feedback adjustment instructions, and send the feedback adjustment instructions to the control execution module to execute the target feedback control scheme.

[0067] Furthermore, the present application S500 includes: A feedback control scheme identifier is pre-built, and the feedback control scheme identifier is used to identify the abnormal correlation causal clue set and the abnormality degree set to obtain the target feedback control scheme.

[0068] Specifically, the Feedback Control Scenario Identifier is a module that automatically identifies and evaluates feedback control strategies. Based on input data (such as a set of abnormal causal clues and a set of abnormality levels), it uses pre-built rules or algorithms to determine the appropriate control strategy. Simply put, the Feedback Control Scenario Identifier generates an appropriate control strategy based on abnormal conditions.

[0069] Collect historical anomaly data, including anomaly causal clues, anomaly severity, and system status data. Process the collected anomaly data to ensure clarity and consistency. This typically involves removing noise, filling missing values, and normalizing numerical ranges. Extract the causal relationships that led to the anomaly from the historical anomaly data. Analyze deviations in the historical data and calculate the severity of each anomaly causal clue to form an anomaly severity set. Develop decision rules for feedback control solutions based on historical data, expert experience, or data analysis models. Validate the established model to ensure it accurately generates appropriate control solutions under different anomaly scenarios.

[0070] The set of anomaly-related causal clues and the set of anomaly degrees are input into the feedback control solution identifier to identify the most appropriate control solution. For example, if an overload condition is detected, accompanied by a high-temperature deviation, the problem needs to be addressed by reducing the load or activating a backup cooling system. The goal of the control solution is to optimize and regulate system operations to restore the system to a normal or optimal state, thereby avoiding excessive energy waste or equipment damage. Once the target feedback control solution is determined, specific operational instructions are generated based on the target feedback control solution, such as adjusting parameters, activating equipment, and adjusting processes. The set of anomaly-related causal clues, the set of anomaly degrees, and the target feedback solution are stored in the control memory and used as a basis for subsequent decarbonization process control. Upon receiving the feedback adjustment instructions, the control execution module adjusts the equipment's operating state accordingly, ensuring the accurate execution of the feedback control solution and effectively resolving the anomaly. SPARK focuses on real-time, data-driven feedback control, utilizing methods such as causal relationship analysis and anomaly degree identification to trigger appropriate control actions. The pre-built feedback control solution identifier automatically adapts to anomalies, adjusting process flows and optimizing operating parameters, embodying SPARK's adaptive adjustment process. Through feedback adjustment instructions, closed-loop optimization can be achieved, deviations in the process flow can be adjusted, and the efficiency and stability of the production process can be ensured.

[0071] By implementing the target feedback control scheme through the control execution module, the expected results are as follows: the typical investment cost of the S module is 5-20 million yuan, with an annual emission reduction benefit of 12,000-30,000 tons of CO2, and an ROI period of 5-8 years; the typical investment cost of the P module is 500,000-3 million yuan, with an annual emission reduction benefit of 3,000-8,000 tons of CO2, and an ROI period of 1-3 years; the typical investment cost of the A+R module is 800,000-5 million yuan, with an annual emission reduction benefit of 5,000-15,000 tons of CO2, and an ROI period of 2-4 years; the typical investment cost of the K module is 200,000-1 million yuan, with an emission reduction benefit of 30% efficiency increase in industry replication, and an ROI period of 1 year. By constructing a feedback control scheme identifier, it automatically generates reasonable control schemes for abnormal conditions, reducing manual intervention. Automatically optimizing control schemes helps reduce energy waste, for example, by automatically adjusting equipment loads or activating backup systems, reducing energy consumption without sacrificing production efficiency.

[0072] In summary, the AI-driven industrial decarbonization process control method based on feedback regulation provided in this application has the following beneficial effects: By executing industrial decarbonization process monitoring, a process operation data flow sequence is obtained, wherein each process operation data flow includes equipment parameters, energy consumption data, carbon emission values ​​and environmental disturbance factors; the process operation data flow sequence is traversed to perform iterative analysis of process operating conditions, and the iterative abnormal operating condition state characteristics are obtained; based on the iterative abnormal operating condition state characteristics, a match is performed in a pre-built causal knowledge base to obtain a set of associated causal clues; the associated causal clue set is traversed to perform historical data backtracing anomaly investigation to obtain an abnormal associated causal clue set and an abnormality degree set; feedback control scheme identification is performed based on the abnormal associated causal clue set and the abnormality degree set to obtain a target feedback control scheme, and a feedback adjustment instruction is generated, and the feedback adjustment instruction is sent to the control execution module to execute the target feedback control scheme. In other words, by collecting the process operation data flow sequence, conducting iterative analysis of the process operating conditions, extracting the state characteristics of abnormal operation, and using the pre-built causal knowledge base to trace the causes of carbon emission anomalies, and through historical data backtracking, the causes of the anomalies are associated with the actual operating conditions, and the target feedback control scheme is determined to dynamically intervene in the decarbonization process, achieving accurate tracing and efficient control of carbon emission anomalies, and greatly improving the control effect of the industrial decarbonization process.

[0073] Example 2, based on the same inventive concept as the AI-driven industrial decarbonization process control method based on feedback regulation in the aforementioned Example 1, this application also provides an AI-driven industrial decarbonization process control system based on feedback regulation, please refer to the attached Figure 2 The AI-driven industrial decarbonization process control system based on feedback regulation includes: The process monitoring module 11 is used to perform industrial decarbonization process monitoring and obtain a process operation data flow sequence, wherein each process operation data flow includes equipment parameters, energy consumption data, carbon emission values ​​and environmental disturbance factors; the timing iteration analysis module 12 is used to traverse the process operation data flow sequence to perform timing iteration analysis of process operating conditions and obtain iterative abnormal operating condition status characteristics; the knowledge base matching module 13 is used to match the pre-built causal knowledge base based on the iterative abnormal operating condition status characteristics to obtain a set of associated causal clues; the anomaly investigation module 14 is used to traverse the associated causal clue set to perform historical data backtracing anomaly investigation and obtain an abnormal associated causal clue set and an abnormal degree set; the control scheme execution module 15 is used to identify the feedback control scheme based on the abnormal associated causal clue set and the abnormal degree set, obtain the target feedback control scheme, and generate feedback adjustment instructions, and send the feedback adjustment instructions to the control execution module to execute the target feedback control scheme.

[0074] Furthermore, the process monitoring module 11 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: A multimodal sensing array is deployed at the key nodes of the target industrial decarbonization process, wherein the multimodal sensing array includes an equipment parameter collector, an energy consumption metering unit, a carbon emission monitoring device and an environmental disturbance perception module; the multimodal sensing array is traversed to perform simultaneous sequential perception of the key nodes to construct a time-synchronized process operation data flow sequence.

[0075] Furthermore, the time sequence iteration analysis module 12 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: Extract a first process operation data flow from the process operation data flow sequence; perform operating condition identification on the first process operation data flow based on a preset process operating condition standard value to obtain a first process operation operating condition state vector, and add the preset process operating condition standard value and the first process operation operating condition state vector into a first iterative memory unit; perform process operating condition time series iterative analysis on the second process operation data flow extracted from the process operation data flow sequence based on the first iterative memory unit to obtain a second process operating condition state vector; update the first iterative memory unit based on the second process operating condition state vector to obtain a second iterative memory unit, and so on, perform time series iterative analysis on the remaining process operation condition state vectors in the process operation data flow sequence to obtain a target iterative memory unit; perform abnormal feature identification on the target iterative memory unit to obtain the iterative abnormal operating condition state feature.

[0076] Furthermore, the time sequence iteration analysis module 12 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: Extract the preset process operating condition standard value from the first iterative memory unit, perform operating condition identification on the second process operation data stream, and obtain a second initial process operation condition state vector; extract the first process operation condition state vector from the first iterative memory unit, perform time series iterative enhancement on the second initial process operation condition state vector, and obtain a second process operation condition state vector.

[0077] Furthermore, the time sequence iteration analysis module 12 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: Calculate the element similarity between the first process operating condition state vector and the second initial process operating condition state vector to obtain an element similarity set; perform matrix processing based on the element similarity set to construct a timing iteration enhancement matrix; use the timing iteration enhancement matrix to perform convolution enhancement on the second initial process operating condition state vector to obtain the second process operating condition state vector.

[0078] Furthermore, the knowledge base matching module 13 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: Obtain a set of historical abnormal related events of the industrial decarbonization process; traverse the set of historical abnormal related events to extract causal clues, and obtain a set of historical abnormal causal clues and a corresponding set of historical abnormal operating condition state features; integrate the set of historical abnormal causal clues and the corresponding set of historical abnormal operating condition state features to obtain an integrated set of historical abnormal causal clues and a corresponding integrated set of historical abnormal operating condition state features; associate and store the integrated set of historical abnormal causal clues and the corresponding integrated set of historical abnormal operating condition state features to obtain the causal knowledge base.

[0079] Furthermore, the knowledge base matching module 13 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: The historical abnormal operating condition state feature set is integrated in the same category, and the integration results are respectively averaged to obtain the integrated historical abnormal operating condition state feature set; based on the integrated historical abnormal operating condition state feature set, the historical abnormal causal clue set is mapped and integrated, and the mapping integration results are unioned to obtain the integrated historical abnormal causal clue set.

[0080] Furthermore, the abnormality troubleshooting module 14 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: In the historical data backtracking window, the associated causal clue set is traversed to perform historical data backtracking to obtain the associated causal clue historical data sequence set; the associated causal clue historical data sequence set is traversed to perform standard value deviation identification to obtain the associated causal clue deviation value sequence set; the associated causal clue set is checked for abnormalities based on the associated causal clue deviation value sequence set to obtain the abnormal associated causal clue set; the abnormal associated causal clue deviation value sequence corresponding to the abnormal associated causal clue set is extracted from the associated causal clue deviation value sequence set to perform deviation value mean processing to obtain the abnormality degree set.

[0081] Furthermore, the control scheme execution module 15 in the AI-driven industrial decarbonization process control system based on feedback regulation is also used to: A feedback control scheme identifier is pre-built, and the feedback control scheme identifier is used to identify the abnormal correlation causal clue set and the abnormality degree set to obtain the target feedback control scheme.

[0082] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The AI-driven industrial decarbonization process control method based on feedback regulation and the specific examples in Example 1 are also applicable to the AI-driven industrial decarbonization process control system based on feedback regulation in this embodiment. Through the above detailed description of the AI-driven industrial decarbonization process control method based on feedback regulation, those skilled in the art can clearly understand the AI-driven industrial decarbonization process control system based on feedback regulation in this embodiment, so for the sake of conciseness of the specification, it will not be described in detail here.

[0083] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0084] Obviously, for those skilled in the art, several improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the scope of protection of the present application.

Claims

1. An AI-driven industrial decarbonization process control method based on feedback regulation, characterized in that: include: Perform industrial decarbonization process monitoring to obtain a sequence of process operation data streams, where each process operation data stream includes equipment parameters, energy consumption data, carbon emission values, and environmental disturbance factors; Traversing the process operation data flow sequence to perform iterative analysis of the process working condition time sequence to obtain the state characteristics of iterative abnormal operating conditions; Based on the iterative abnormal operating condition status characteristics, the pre-built causal knowledge base is matched to obtain a set of related causal clues; Traversing the associated causal clue set to perform historical data backtracking anomaly investigation, and obtaining an abnormal associated causal clue set and an abnormality degree set; Based on the abnormality-related causal clue set and the abnormality degree set, feedback control scheme identification is performed to obtain a target feedback control scheme, and feedback adjustment instructions are generated. The feedback adjustment instructions are sent to the control execution module to execute the target feedback control scheme.

2. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 1 is characterized in that: Perform industrial decarbonization process monitoring to obtain a sequence of process operation data streams. Each process operation data stream includes equipment parameters, energy consumption data, carbon emission values, and environmental disturbance factors, including: Deploy a multimodal sensing array at key nodes of the target industrial decarbonization process, wherein the multimodal sensing array includes an equipment parameter collector, an energy consumption metering unit, a carbon emission monitoring device, and an environmental disturbance sensing module; The multimodal sensing array is traversed to perform simultaneous sequential sensing of key nodes, and a time-synchronized process running data flow sequence is constructed.

3. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 1 is characterized in that: Traverse the process operation data flow sequence to perform iterative analysis of the process operating condition time sequence to obtain the iterative abnormal operating condition status characteristics, including: Extracting a first process running data flow from the process running data flow sequence; Identify the working condition of the first process operation data stream based on a preset process working condition standard value to obtain a first process working condition state vector, and add the preset process working condition standard value and the first process working condition state vector to a first iterative memory unit; Performing a process operating condition time series iterative analysis on the second process operation data stream extracted from the process operation data stream sequence based on the first iterative memory unit to obtain a second process operation condition state vector; The first iterative memory unit is updated based on the second process operation condition state vector to obtain a second iterative memory unit, and so on, the remaining process operation condition state vectors in the process operation data flow sequence are subjected to time series iterative analysis to obtain a target iterative memory unit; Abnormal feature identification is performed on the target iterative memory unit to obtain the abnormal operating condition state feature of the iterative operation.

4. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 3 is characterized in that: Performing a process condition time series iterative analysis on the second process operation data flow extracted from the process operation data flow sequence based on the first iterative memory unit to obtain a second process operation condition state vector includes: Extracting a preset process operating condition standard value from the first iterative memory unit, performing operating condition identification on the second process operation data stream, and obtaining a second initial process operation condition state vector; A first process operation condition state vector is extracted from the first iterative memory unit, and the second initial process operation condition state vector is subjected to time series iterative enhancement to obtain a second process operation condition state vector.

5. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 4 is characterized in that: Extracting a first process operation condition state vector from the first iterative memory unit, performing time sequence iterative enhancement on the second initial process operation condition state vector, and obtaining a second process operation condition state vector, including: Calculating element similarities between the first process operation condition state vector and the second initial process operation condition state vector to obtain an element similarity set; Performing matrix processing based on the element similarity set to construct a time series iterative enhancement matrix; The second initial process operating condition state vector is convolution-enhanced using the timing iteration enhancement matrix to obtain the second process operating condition state vector.

6. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 1 is characterized in that: Based on the iterative abnormal operating condition status characteristics, the pre-built causal knowledge base is matched to obtain a set of related causal clues, including: Obtain a collection of historical abnormal correlation events of the industrial decarbonization process; Traversing the historical abnormality associated event set to extract causal clues, and obtaining a historical abnormality causal clue set and a corresponding historical abnormal operating condition state feature set; Integrating the historical abnormal causal clue set and the corresponding historical abnormal operating condition state feature set to obtain an integrated historical abnormal causal clue set and a corresponding integrated historical abnormal operating condition state feature set; The integrated historical abnormal causal clue set and the corresponding integrated historical abnormal operating condition state feature set are associated and stored to obtain the causal knowledge base.

7. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 6 is characterized in that: Integrating the historical abnormal causal clue set and the corresponding historical abnormal operating condition state feature set to obtain an integrated historical abnormal causal clue set and the corresponding integrated historical abnormal operating condition state feature set, including: Performing similar integration on the historical abnormal operating condition state feature set, and performing mean processing on the integration results respectively, to obtain an integrated historical abnormal operating condition state feature set; The historical abnormal causal clue set is mapped and integrated based on the integrated historical abnormal operating condition state feature set, and the mapping and integration results are unioned to obtain the integrated historical abnormal causal clue set.

8. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 1 is characterized in that: Traverse the associated causal clue set to perform historical data backtracking anomaly investigation to obtain an abnormal associated causal clue set and an abnormality degree set, including: In the historical data backtracking window, traverse the associated causal clue set to perform historical data backtracking to obtain an associated causal clue historical data sequence set; Traversing the associated causal clue historical data sequence set to identify standard value deviations, and obtaining an associated causal clue deviation value sequence set; Performing abnormality screening on the associated causal clue set based on the associated causal clue deviation value sequence set to obtain an abnormal associated causal clue set; The abnormal associated causal clue deviation value sequence corresponding to the abnormal associated causal clue set is extracted from the associated causal clue deviation value sequence set, and deviation value mean processing is performed to obtain the abnormal degree set.

9. The AI-driven industrial decarbonization process control method based on feedback regulation according to claim 1 is characterized in that: A feedback control scheme identifier is pre-built, and the feedback control scheme identifier is used to identify the abnormal correlation causal clue set and the abnormality degree set to obtain the target feedback control scheme.

10. AI-driven industrial decarbonization process control system based on feedback regulation, characterized by: Steps for implementing the AI-driven industrial decarbonization process control method based on feedback regulation according to any one of claims 1 to 9, wherein the AI-driven industrial decarbonization process control system based on feedback regulation comprises: The process monitoring module is used to perform industrial decarbonization process monitoring and obtain a sequence of process operation data streams, where each process operation data stream includes equipment parameters, energy consumption data, carbon emission values, and environmental disturbance factors; A time series iteration analysis module is used to traverse the process operation data flow sequence to perform time series iteration analysis of the process working condition and obtain the state characteristics of the iterative abnormal working condition; The knowledge base matching module is used to match the pre-built causal knowledge base based on the iterative abnormal operating condition status characteristics to obtain a set of related causal clues; An anomaly troubleshooting module, configured to traverse the associated causal clue set to perform historical data backtracking anomaly troubleshooting and obtain an abnormal associated causal clue set and an abnormality degree set; The control scheme execution module is used to identify the feedback control scheme based on the abnormal association causal clue set and the abnormality degree set, obtain the target feedback control scheme, and generate feedback adjustment instructions, and send the feedback adjustment instructions to the control execution module to execute the target feedback control scheme.

Images