A method and system for intelligent monitoring and early warning of operating conditions of a VOCs incinerator

Abstract

This invention provides a method and system for intelligent monitoring and early warning of VOCs incinerator operating conditions. The method includes: collecting operating condition data such as combustion chamber temperature, system pressure, inlet and outlet VOCs concentrations, key actuator status, and bypass valve status based on edge monitoring nodes deployed on-site; aligning the sampling period of the multi-source data and generating timestamps to form a time-ordered original operating condition data stream; filtering key operating data from the data stream and encapsulating it into key operating data frames; calculating derived indicators such as purification efficiency; generating data integrity identification information corresponding one-to-one with the key operating data frames; further constructing a tamper-proof evidence chain containing chained summaries to achieve traceable verification of the continuity of operating data; and, under the premise that the evidence chain consistency verification is passed, constructing a consistency feature vector and calculating the consistency deviation, and introducing an ant colony algorithm on the operating mode diagram for multi-path consistency evaluation.

Description

Technical Field

[0001] This invention relates to the field of industrial waste gas treatment and environmental monitoring technology, and in particular to a method and system for intelligent monitoring and early warning of VOCs incinerator operating conditions based on reinforcement learning and ant colony algorithm. Background Technology

[0002] Volatile organic compound (VOCs) incinerators are crucial equipment for industrial waste gas treatment, widely used in industries such as chemical, pharmaceutical, coating, and electronics manufacturing. Their operational stability and treatment effectiveness directly impact pollutant emission compliance and production safety. Existing VOCs incinerators typically monitor basic operations through parameters such as combustion chamber temperature, pressure, and inlet / outlet concentrations, triggering alarms when these parameters exceed limits to prompt maintenance personnel for intervention. However, these methods generally suffer from the following shortcomings: First, existing monitoring methods mostly rely on fixed thresholds or single-parameter alarm mechanisms, which have limited adaptability to fluctuations in operating conditions, load changes, and seasonal differences. They are prone to false alarms under normal fluctuation conditions or missed alarms in complex scenarios such as hidden degradation and bypassing, making it difficult to identify potential violations or abnormal operating states in a timely manner. Second, existing systems mostly process multi-source operating data at the level of "real-time display or simple storage," lacking systematic analysis of the inherent logical consistency between data, making it difficult to determine whether the incinerator is in a truly effective governance state from the perspective of the overall operating mechanism.

[0003] Therefore, we propose an intelligent monitoring and early warning method and system for the operating conditions of VOCs incinerators. The information disclosed in the background section is only for enhancing the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a method and system for intelligent monitoring and early warning of VOCs incinerator operating conditions, thereby solving the technical problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for intelligent monitoring and early warning of VOCs incinerator operating conditions includes the following steps: S1. The edge monitoring node collects operating condition data from the VOCs incinerator operating system in real time. The operating condition data includes at least: incinerator combustion chamber temperature data, system pressure data, inlet and outlet VOCs concentration data, operating status data of key actuators and bypass valve status data. A unified timestamp is added to the collected operating condition data to form a time-ordered original operating condition data stream. S2. In the original operating condition data stream, select key operating data that reflects the effectiveness of VOCs incinerator pollution control and operational compliance, and generate data integrity identification information for the key operating data in chronological order, so that the key operating data at each moment and its corresponding data integrity identification information form a one-to-one correspondence. S3. Based on key operational data and its corresponding data integrity identification information, construct a continuous and related anti-tampering evidence chain in chronological order, so that the key operational data at the current moment and the key operational data at the previous moment form an irreversible logical relationship, thereby achieving anti-tampering protection for the continuity of key operational data. S4. Utilize the tamper-proof evidence chain to verify the consistency of key operating data. By analyzing the logical consistency between inlet and outlet VOCs concentration data, bypass valve status data, and operating status data of key actuators, identify abnormal operating states that are inconsistent with the normal operating mechanism and determine the abnormal operating states as potential violations or abnormal working conditions. S5. When an abnormal operating state is detected, the abnormal operating condition evidence solidification process is triggered. The tamper-proof evidence chain fragment corresponding to the abnormal operating state is encrypted, encapsulated, and stored as abnormal operating condition evidence data. At the same time, the corresponding level of operating condition warning information is generated according to the degree of abnormality. S6. Output the operating condition warning information locally and report it remotely. During the reporting process, upload the integrity verification information of the abnormal operating condition evidence data simultaneously, so that the receiving end can perform consistency verification and traceability review of the operating condition warning information and its corresponding abnormal operating condition evidence data.

[0006] S1 specifically involves: establishing data acquisition connections with the incinerator control system and field sensors through edge monitoring nodes deployed at the VOCs incinerator site to acquire operational condition data acquisition channels reflecting the incinerator's operating status; collecting in real-time data from the operational condition data acquisition channels, including incinerator combustion chamber temperature, system pressure, inlet and outlet VOCs concentration, operational status data of key actuators, and bypass valve status data, according to a unified sampling period, to form raw operational condition data; assigning a timestamp to each raw operational condition data point under a unified time reference to ensure sequential consistency of the raw operational condition data in the time dimension; and combining the raw operational condition data with the added timestamps into a time-ordered raw operational condition data stream according to the acquisition time sequence, which serves as input data for subsequent key operational data filtering.

[0007] S2 specifically involves: based on the pollution control mechanism of VOCs incinerators, selecting key operational data from the original operating condition data stream to characterize the compliance and effectiveness of incinerator operation; key operational data includes at least inlet and outlet VOCs concentration data directly related to pollutant removal, bypass valve status data directly related to the operating status of the treatment facilities, and combustion chamber temperature data related to the combustion process; for each selected key operational data, generating data integrity identification information according to its corresponding timestamp, so that the key operational data and data integrity identification information form a one-to-one correspondence; and outputting the key operational data and its data integrity identification information that form a correspondence as the basic data unit for constructing an anti-tampering evidence chain.

[0008] S3 specifically involves: arranging the basic data units sequentially according to timestamp order to determine the temporal relationship between key operational data at adjacent times; associating the key operational data at the current time and its data integrity identifier with the integrity identifier of the key operational data at the previous time to generate a continuous association relationship; based on the continuous association relationship, constructing an anti-tampering evidence chain covering a continuous time period, so that the key operational data form an irreversible time sequence association in logical structure; and using the constructed anti-tampering evidence chain as a trusted data foundation for subsequent operational consistency verification and anomaly identification.

[0009] S4 specifically involves: based on an anti-tampering evidence chain, performing operational consistency verification on key operational data over a continuous time period, and analyzing the logical relationships between inlet and outlet VOCs concentration data, bypass valve status data, and key actuator operational status data; during the operational consistency verification process, an ant colony algorithm is introduced to construct a multi-path consistency evaluation model, which comprehensively evaluates the consistency matching degree of the current key operational data under different paths by simulating various possible normal operation paths; combined with a reinforcement learning mechanism, the parameters of the consistency evaluation model are adaptively updated, enabling the model to continuously strengthen its ability to identify normal operation modes based on historical operational results; when the matching degree of the current key operational data in the consistency evaluation model is continuously lower than the preset consistency threshold, the corresponding operational status is identified as an abnormal operational status, and an abnormal operational status identifier is output.

[0010] S5 specifically involves: upon identifying an abnormal operating state, triggering the abnormal operating condition evidence solidification process; extracting evidence chain fragments corresponding to the abnormal operating state from the tamper-proof evidence chain; encrypting and encapsulating the evidence chain fragments to form abnormal operating condition evidence data; storing the abnormal operating condition evidence data in the protected storage area of ​​the edge monitoring node; determining the warning level of the abnormal operating state based on its duration, frequency, and consistency deviation, combined with the risk assessment results obtained from reinforcement learning; generating operating condition warning information corresponding to the warning level; and associating and marking the operating condition warning information with the abnormal operating condition evidence data.

[0011] Specifically, S6 involves: outputting operational condition warning information locally at the VOCs incinerator site in a visual manner to alert on-site maintenance personnel to the current abnormal operating status; sending operational condition warning information to a remote monitoring or management platform via remote communication; simultaneously uploading integrity verification information of associated abnormal operating condition evidence data to support remote verification of the authenticity of the abnormal operating condition evidence data; and enabling the remote receiving end to perform consistency verification and traceability review of the operational condition warning information and its corresponding abnormal operating condition evidence data based on the integrity verification information, thus forming a complete closed loop of warning and evidence collection.

[0012] A VOCs incinerator operating condition intelligent monitoring and early warning system includes: Edge monitoring nodes are deployed at the VOCs incinerator site to establish data acquisition connections with the incinerator control system and on-site sensors, and are configured with multiple operating condition data acquisition channels. The key operational data processing module, located in the edge monitoring node, is used to filter key operational data that characterizes the effectiveness of VOCs incinerator treatment and operational compliance from the time-ordered raw operational data stream; The data integrity identifier generation module is set in the edge monitoring node and is used to generate data integrity identifier information that corresponds to each key running data frame. The tamper-proof evidence chain construction module is set in the edge monitoring node and is used to perform sequential consistency verification on key operational data frames at consecutive times based on the timestamp field and sequence number field of key operational data frames. The consistency assessment and anomaly identification module is set up in the edge monitoring node to construct a feature vector reflecting the consistency of the operating mechanism, provided that the consistency verification of the anti-tampering evidence chain is passed. The abnormal operating condition evidence solidification and early warning generation module is set in the edge monitoring node. It is used to extract evidence chain fragments containing data before and after the occurrence of the abnormal operating state from the anti-tamper evidence chain when an abnormal operating state is identified. The early warning output and remote reporting module is set in the edge monitoring node and is used to output local early warning information at the VOCs incinerator site, including abnormal event identifier, early warning level, risk score and evidence time range; The remote consistency verification module is set up in a remote monitoring or management platform to establish a one-to-one correspondence between early warning information and abnormal operating condition evidence data based on abnormal event identifiers and evidence package summaries.

[0013] The beneficial effects of this invention are as follows: This invention aligns the sampling periods and timestamps multi-source operational data to form a time-ordered operational data stream, effectively avoiding misjudgments caused by asynchronous multi-source data. This ensures the comparability and analytical accuracy of different parameters at the same time scale, providing a reliable foundation for subsequent consistency assessments. By generating data integrity identifiers for key operational data and constructing a chain-like summary of tamper-proof evidence, any operational data at any given time is irreversibly linked to preceding and following data. Any data tampering can be detected, thus achieving integrity protection and traceability verification of operational data, significantly enhancing regulatory evidence collection and accountability capabilities.

[0014] This invention constructs a consistency feature vector encompassing inlet and outlet VOCs concentrations, purification efficiency, bypass valve status, combustion chamber temperature, and actuator status, and calculates the consistency deviation. This allows for a holistic assessment of the incinerator's operating status from an operational mechanism perspective, identifying hidden anomalies where individual parameters have not exceeded limits but the overall system has deviated from normal operating conditions. The invention utilizes an ant colony algorithm for multi-path consistency evaluation on the operating mode diagram, avoiding the limitations of single rules or fixed models. Furthermore, reinforcement learning is combined to adaptively update the consistency evaluation parameters, enabling the system to continuously optimize early warning criteria under different loads, operating conditions, and long-term operation, thereby reducing false alarm rates and improving the stability of anomaly identification.

[0015] This invention achieves real-time solidification of abnormal data by simultaneously capturing fragments of the evidence chain before and after the anomaly occurs and constructing encrypted abnormal operating condition evidence data packets when identifying abnormal operating states. Simultaneously, it generates tiered early warning information based on risk scores, enabling maintenance personnel and regulatory departments to take differentiated measures according to risk levels, thus improving emergency response and management efficiency. By binding early warning information to abnormal operating condition evidence data one-to-one and employing a hierarchical reporting, receipt confirmation, and remote consistency recalculation verification mechanism, data is guaranteed not to be lost or confused even under network fluctuations or interruptions. This achieves a complete closed loop of operating condition early warning and evidence review, enhancing the system's reliability and practicality in real-world industrial environments. Attached Figure Description

[0016] Figure 1This is a schematic diagram of an intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to the present invention; Figure 2 This is a schematic diagram of the framework of an intelligent monitoring and early warning system for the operating conditions of a VOCs incinerator according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1: As Figure 1 As shown in the figure, this embodiment provides a method for intelligent monitoring and early warning of the operating conditions of a VOCs incinerator, including the following steps: S1. Operational condition data acquisition steps: The edge monitoring node collects operational condition data in real time from the VOCs incinerator's operating system. The operational condition data includes at least: incinerator combustion chamber temperature data, system pressure data, inlet and outlet VOCs concentration data, operating status data of key actuators, and bypass valve status data. A unified timestamp is added to the collected operational condition data to form a time-ordered original operational condition data stream. S2. Key Operational Data Screening and Integrity Identification Steps: In the original operational data stream, select key operational data that reflects the effectiveness of VOCs incinerator pollution control and operational compliance, and generate data integrity identification information for the key operational data in chronological order, so that the key operational data at each moment and its corresponding data integrity identification information form a one-to-one correspondence. S3. Steps for constructing an anti-tampering evidence chain: Based on key operational data and its corresponding data integrity identification information, construct a continuous and related anti-tampering evidence chain in chronological order, so that the key operational data at the current moment and the key operational data at the previous moment form an irreversible logical relationship, thereby achieving anti-tampering protection for the continuity of key operational data. S4. Operational Consistency Verification and Anomaly Identification Steps: Using the anti-tampering evidence chain, the operational consistency of key operational data is verified. By analyzing the logical consistency relationship between inlet and outlet VOCs concentration data, bypass valve status data, and operational status data of key actuators, abnormal operational states that are inconsistent with the normal operation mechanism are identified, and the abnormal operational states are judged as potential violations or abnormal working conditions. S5. Abnormal operating condition evidence solidification and graded early warning steps: When an abnormal operating state is identified, the abnormal operating condition evidence solidification process is triggered. The tamper-proof evidence chain fragments corresponding to the abnormal operating state are encrypted, encapsulated, and stored as abnormal operating condition evidence data. At the same time, operating condition early warning information of the corresponding level is generated according to the degree of abnormality. S6. Early warning output and traceable reporting steps: Output the operating condition early warning information locally and report it remotely. During the reporting process, upload the integrity verification information of the abnormal operating condition evidence data simultaneously, so that the receiving end can perform consistency verification and traceability review on the operating condition early warning information and its corresponding abnormal operating condition evidence data.

[0019] S1 specifically includes the following sub-steps: S110. Steps for establishing and mapping operating condition data acquisition channels: By deploying edge monitoring nodes at the VOCs incinerator site, data acquisition connections are established with the incinerator control system and field sensors to form operating condition data acquisition channels for acquiring operating condition data. The operating condition data acquisition channels include at least temperature acquisition channels, pressure acquisition channels, concentration acquisition channels, status acquisition channels, and valve position acquisition channels. Among them, the edge monitoring nodes configure corresponding operating condition data acquisition channels for each type of operating condition data and establish channel-data item mapping relationships so that each operating condition data acquisition channel can uniquely correspond to its output data type and data item name. The data acquisition connection can adopt any one or more of the following implementation methods: fieldbus, industrial Ethernet, serial communication, or analog acquisition modules, to ensure that the edge monitoring nodes can stably acquire the data sources required for subsequent acquisition.

[0020] S120. Unified Sampling Period Acquisition and Time Grid Alignment Steps: Based on the channel-data item mapping relationship, the edge monitoring node performs polling or synchronous sampling of the operating condition data acquisition channels according to the unified sampling period, acquiring and forming raw operating condition data in real time. The raw operating condition data includes at least: incinerator combustion chamber temperature data, system pressure data, inlet and outlet VOCs concentration data, operating status data of key actuators, and bypass valve status data. When different operating condition data acquisition channels have asynchronous outputs, the edge monitoring node aligns the asynchronously output data to a unified time grid. The unified time grid satisfies the following: in, To standardize the sampling period, To align the start time, For the first Each aligned sampling time, It is a non-negative integer; by aligning the original operating condition data to a unified time grid, the data processed in subsequent steps remains comparable and consistent on the time scale.

[0021] S130, Unified Time Base Timestamp Generation and Synchronization Error Constraint Steps: The edge monitoring node generates timestamp information for each piece of original operating condition data based on the unified time base, and binds the timestamp information with the corresponding original operating condition data; the edge monitoring node performs periodic time synchronization with the unified time base to suppress clock drift, ensuring that the synchronization error of the timestamp information relative to the unified time base meets the following requirements: in, For synchronization error, The maximum allowable synchronization error threshold is set; by constraining the synchronization error, the original operating condition data is made consistent and traceable in the time dimension; in an optional implementation, periodic time synchronization can be achieved through network time synchronization or satellite time synchronization to further improve the stability of the unified time reference.

[0022] S140, Standard Data Frame Encapsulation, Quality Flag Marking, and Ordered Buffer Output Steps: The edge monitoring node encapsulates the raw operating condition data with added timestamp information into a standard data frame. The standard data frame includes at least a timestamp field, a data type field, a data value field, a data unit field, and a data quality flag field. The data quality flag field is used to indicate abnormal sampling states such as over-range, missing, and jump. The edge monitoring node writes the standard data frame into a time-ordered buffer queue according to the timestamp information and outputs it in chronological order to form a time-ordered raw operating condition data stream. When missing or abnormal sampling occurs, the edge monitoring node marks the corresponding standard data frame according to the data quality flag field, thereby ensuring the availability and interpretability of the input data for subsequent key operating data filtering steps.

[0023] S2 specifically includes the following sub-steps: S210, Determination and Extraction of the Minimum Set of Key Operating Data: The edge monitoring node takes the time-ordered raw operating condition data stream output by S140 as input. Based on the requirements for judging the effectiveness of VOCs incinerator pollution control and operational compliance, it determines and extracts key operating data from the time-ordered raw operating condition data stream. The key operating data is the minimum data set used to characterize the effectiveness and compliance of the control. The minimum data set includes at least: inlet VOCs concentration data, outlet VOCs concentration data, bypass valve status data, combustion chamber temperature data, and operating status data of key actuators. Among them, each data item uses the data value field and data unit field of the standard data frame to ensure that the same data item has a consistent unit caliber at different sampling times, and to enable subsequent steps to perform comparable analysis on the same data item.

[0024] S220, Key Operational Data Frame Encapsulation and Purification Efficiency Derived Data Calculation Steps: Following the same encapsulation rules as S140, the edge monitoring node encapsulates the key operational data corresponding to each aligned sampling time into a key operational data frame. The key operational data frame includes at least: a timestamp field, a data type field, a data value field, a data unit field, a data quality flag field, and a sequence number field. The sequence number field indicates the position of the key operational data frame in the time series. Based on this, the edge monitoring node calculates the purification efficiency data based on the influent VOCs concentration data and effluent VOCs concentration data within the same key operational data frame. It satisfies: in, For the first VOCs concentration data at each aligned sampling time point For the first VOCs concentration data at each aligned sampling time point To prevent the minimum concentration lower limit constant from having a denominator of zero, For the first Purification efficiency data at each aligned sampling time; To maximize the function; the purification efficiency data is written into the key operation data frame as a derived data item of the key operation data, and is associated with the bypass valve status data, combustion chamber temperature data and the operation status data of the key actuators at the same time with a timestamp consistent relationship.

[0025] S230. Data Integrity Identifier Generation Steps: The edge monitoring node generates a one-to-one data integrity identifier for each critical operational data frame. The data integrity identifier is a hash digest obtained by performing digest calculation on the critical operational data frame. The hash digest satisfies: in, For the preset hash function, For the first The key running data frames at each aligned sampling moment; For hash digest; Sampling sequence number; key running data frame The hash digest is obtained by concatenating at least the timestamp field, data type field, data value field, data unit field, data quality flag field, and sequence number field in a preset order. With critical runtime data frames This establishes a one-to-one correspondence and enables key operational data frames at different sampling times to be uniquely located through the sequence number field and the timestamp field.

[0026] S240, Key Operational Data and Integrity Identifier Output and Abnormal Continuity Assurance Steps: The edge monitoring node outputs each key operational data frame with its corresponding hash digest. The binding process creates a one-to-one correspondence between "critical runtime data frame - data integrity identification information" output unit, which is then continuously output in timestamp order to subsequent steps in constructing the anti-tampering evidence chain. When the data quality flag field of the critical runtime data frame indicates a missing, abrupt, or out-of-range condition, the edge monitoring node still generates a hash digest corresponding to that moment. Furthermore, the data quality flag field and placeholder data values ​​are retained in the critical runtime data frames to ensure the accuracy of the critical runtime data frames and their corresponding hash digests. Continuous output in time series prevents subsequent steps in building the anti-tampering evidence chain from being broken due to data gaps, thereby improving the integrity and traceability of the evidence chain construction.

[0027] S3 specifically includes the following sub-steps: S310, Key Operational Data Frame Sequence Consistency Verification and Evidence Chain Segmentation Determination Steps: The edge monitoring node takes the "Key Operational Data Frame - Data Integrity Identification Information" output from S240 as input, determines the temporal sequence relationship between adjacent key operational data frames according to the sequence number field and timestamp field of the key operational data frame, and performs a sequence consistency verification to ensure that the sequence number field monotonically increases and the timestamp field does not go backward; when a discontinuous sequence number, a backward timestamp, or a timestamp interval between adjacent key operational data frames is detected, the node will perform a sequence consistency verification. At that time, the edge monitoring node will determine the current key running data frame as the starting point of the new evidence chain segment and generate new chain segment header information to ensure the continuity and verifiability of the subsequent anti-tampering evidence chain in the time dimension.

[0028] S320, Chained Digest Association Calculation and Continuous Association Generation Steps: Edge monitoring nodes generate hash digests based on S230. Correlation calculations are performed between key operational data at adjacent time points to generate a chain-like summary representing continuous correlations. Chained summary By hashing the current time Chained summary from the previous time step By concatenating and summarizing again, the chained summary is obtained, which satisfies the following: in, For the preset hash function, This indicates a concatenation operation performed in a preset order. This refers to the timestamp information corresponding to the timestamp field in the critical runtime data frame. This refers to the sampling sequence number corresponding to the sequence number field in the key running data frame; through chained summaries Also depends on the hash digest at the current time. Chained summary from the previous time step This ensures that any change in key operational data at any given time will affect subsequent chained summaries, thus achieving an irreversible continuous association.

[0029] S330, Segment Start Chain Summary Generation and Segment Header Information Construction Steps: For the initial key running data frame of each evidence chain segment, the edge monitoring node generates a random start amount for the chain segment. And based on the hash digest corresponding to the starting key running data frame. Random starting quantity of chain segment Generate a starting chained digest The initial chained digest satisfies: in, This is the random starting value for the chain segment; A hash digest of the critical running data frame at the beginning of the chain segment; This is a chained digest starting with a segment; the random starting amount for the segment. With the starting chain summary Together, they constitute the head information of the evidence segment and, together with the subsequent chain summary of that evidence segment, form the chain header information. Maintain associated storage to ensure that each piece of evidence has a verifiable starting point and that the summary calculation caliber is consistent with the continuous associations generated by S320.

[0030] S340. The steps for solidifying, outputting, and verifying the consistency of the tamper-proof evidence chain node structure are as follows: The edge monitoring node binds the chained digests generated in chronological order with the corresponding key operational data frames, constructing a tamper-proof evidence chain formed by connecting multiple evidence chain nodes in chronological order; wherein each evidence chain node includes at least one key operational data frame. Corresponding hash digest Corresponding chained summary and chained summaries for characterizing forward associations When outputting a tamper-proof evidence chain, the edge monitoring node must output at least the key runtime data frame. Hash digest With chained summaries It supports traceable verification according to the following consistency verification rules: based on key runtime data frames. Recalculate the hash digest and compare it with the output hash digest. Compare, and then use hash digest Chained summary from the previous time step Recalculate to obtain the chain digest and compare it with the output chain digest. Comparison; when any comparison is inconsistent, it is determined that the corresponding evidence chain node or its forward associated node is at risk of being tampered with, thereby achieving tamper-proof protection and traceability verification of the continuity of key operational data.

[0031] S4 specifically includes the following sub-steps: S410, the steps for constructing consistency features and calculating consistency deviation of trusted input gating: The edge monitoring node takes the evidence chain nodes in the tamper-proof evidence chain output by S340 as input, and first performs consistency verification on the evidence chain nodes. The consistency verification includes at least: the key running data frames. Recalculate the hash digest and merge it with the corresponding hash digest Comparison, and hash digest Recalculate the chained digest with the chained digest from the previous time step and then merge it with the corresponding chained digest. Comparison; only when the consistency verification passes will the edge monitoring node release the corresponding critical runtime data frame. As a trusted input for consistency verification; based on this, the edge monitoring node targets the first... Construct a consistent feature vector at each aligned sampling time. Consistent feature vector At least include: VOCs concentration data VOCs concentration data Purification efficiency data Bypass valve status data Combustion chamber temperature data and the sub-vectors of operational status data of key actuators ; and based on consistent feature vectors Calculate consistency deviation The consistency deviation satisfies: in, For the first A function for the degree of violation of a consistency constraint. For the first The weighting coefficients of each consistency constraint. Number of consistency constraints; consistency deviation. It is used to characterize the degree of deviation of key operational data from normal operating mechanisms at the current moment.

[0032] S420. Ant colony algorithm multi-path consistency evaluation steps based on operation mode graph: Edge monitoring nodes in the preset operation mode graph... The ant colony algorithm is executed to construct a multipath consensus evaluation model, where the set of nodes... Characterizing the operating mode status of a VOCs incinerator, the operating mode status includes at least one or more of the following: ignition, heating, steady state, switching, and shutdown. Characterizes the permissible transitions between operational modes; edge monitoring nodes use a sequence of consistency deviations within a sliding window. As a basis for path evaluation, candidate paths Calculate path cost The path cost satisfies: in, The set of sampling numbers within the sliding window; the ant colony algorithm minimizes path cost. For the target in the running mode diagram Search for the optimal consistent path And the optimal consistency path The corresponding path cost is used to characterize the degree of matching of the current critical operating data under the normal operating path.

[0033] S430. Adaptive update steps for consistency evaluation parameters based on reinforcement learning: The edge monitoring node introduces a reinforcement learning mechanism to adaptively update the consistency evaluation parameters, which at least include a set of consistency constraint weight coefficients. Matching degree threshold One or more of the following; edge monitoring nodes construct reinforcement learning states. Enhance learning state At least include consistency deviation statistics within the sliding window and bypass valve status data. With purification efficiency data The combined features are used to define the actions used to adjust the consistency evaluation parameters as reinforcement learning actions. Edge monitoring nodes generate rewards based on the early warning output and post-verification results. False positives and false negatives correspond to negative rewards, while correct identification corresponds to positive rewards; edge monitoring nodes update the action value function according to the Q-learning update rule. : in, For learning rate, The discount factor is used; through reinforcement learning updates, the consistency assessment parameters adaptively converge under different seasonal loads and VOCs fluctuation conditions, thereby reducing false alarms and improving the ability to identify hidden abnormal operating states. This represents a candidate action in the action space.

[0034] S440. Consistency matching degree calculation and anti-jitter abnormal operation status judgment steps: The edge monitoring node will select the optimal consistency path. Path cost Mapped to consistency matching degree The consistency matching degree satisfies: in, A larger value indicates a higher degree of matching between the current critical operational data and the normal operational path; edge monitoring nodes set matching degree thresholds. And set the sliding window length Number of consecutive judgments When the sliding window length Next continuum Each sliding window satisfies When the corresponding operating status is identified as an abnormal operating status, an abnormal operating status identifier is output, so that the abnormal operating status determination has anti-jitter capability and avoids false alarms caused by single-point noise.

[0035] S5 specifically includes the following sub-steps: S510, Steps for determining the abnormal working condition evidence extraction window and extracting evidence chain fragments: When the... After identifying an abnormal operating state and outputting an abnormal operating state identifier at each aligned sampling moment, the edge monitoring node triggers the abnormal operating condition evidence solidification process, using the sampling sequence number corresponding to the abnormal operating state. Determine the evidence interception window at the center. ,in The length of the front window. The length of the post-window; within the evidence extraction window, the edge monitoring node extracts the evidence chain segment corresponding to the abnormal operating state from the tamper-proof evidence chain output by S340. The evidence chain fragment satisfies: in, For critical running data frames, For hash digest, It is a chain-like summary; through the definition of the evidence capture window, key operational data before and after the occurrence of abnormal operating states can be synchronously solidified into traceable evidence.

[0036] S520, Abnormal Operating Condition Evidence Data Packet Construction, Digest Calculation, and Encryption Encapsulation and Storage Steps: Edge monitoring nodes will construct evidence chain fragments. Encapsulated as an abnormal operating condition evidence data packet Abnormal operating condition evidence data package At least include: abnormal event identifiers The start timestamp corresponding to the evidence capture window With end timestamp Fragments of the chain of evidence And the chain head information of the chain segment to which the evidence chain fragment belongs (including the starting chain summary of the corresponding chain segment); edge monitoring nodes' abnormal operating condition evidence data packets Computational evidence package summary The evidence package summary satisfies: in, A preset hash function is used; based on this, the edge monitoring nodes employ event keys. Abnormal operating condition evidence data package Encrypt and encapsulate the data to form encrypted evidence. The encrypted evidence data satisfies: in, For event keys The symmetric encryption function for the key; event key. From the master key With abnormal event identifier Generated through a key derivation function, satisfying: Edge monitoring nodes will encrypt evidence data. Stored in a protected storage area, and based on anomaly event identifiers. Encrypted evidence data A unique index is used to ensure the confidentiality, integrity, and traceability of evidence data related to abnormal operating conditions.

[0037] S530, Risk Score Calculation, Early Warning Classification and Anti-jitter Output Steps: The edge monitoring node calculates the risk score based on the persistence and severity of the abnormal operating state. Risk score At least by duration indicators Frequency of occurrence index Consistency Deviation Indicators and the risk assessment value output by reinforcement learning Together, we determine that the following conditions must be met: in, , , , Weighting coefficients; duration indicators The frequency index is the number of sliding windows or the duration of abnormal operating conditions. The consistency deviation index is the number of times abnormal operating states are triggered within a preset statistical window. This is a statistic measuring the consistency deviation within a preset statistical window; edge monitoring nodes are based on risk scores. With the set of early warning classification thresholds Determine the warning level: When It was determined to be a Level 1 warning, when It was determined to be a Level II warning, when It was determined to be a Level 3 warning, when It has been determined to be a Level 4 warning; and a limit has been set for the number of consecutive warnings. Only when consecutive When all sliding windows meet the corresponding warning level conditions, the operating condition warning information corresponding to that warning level is output to suppress warning jitter caused by noise and improve warning stability.

[0038] S540, Early Warning Information and Evidence Data Association Marking and One-to-One Correspondence Closed-Loop Steps: When generating operational condition early warning information, the edge monitoring node will mark abnormal events. Summary of Evidence Package The start timestamp corresponding to the evidence capture window With end timestamp The warning level is also written into the operational condition warning information, so that the operational condition warning information can be identified through abnormal event identifiers. Encrypted evidence data in protected storage areas Achieve one-to-one correspondence; and enable the remote receiving end to perform data processing based on the evidence packet digest. Encrypted evidence data Corresponding plaintext evidence data packet Perform consistency checks to form a traceable closed loop of "early warning output - evidence location - integrity check".

[0039] S6 specifically includes the following sub-steps: S610, Local Output and Evidence Location Prompt Steps for Operational Condition Warning Information: The edge monitoring node takes the operational condition warning information generated by S540 as input and performs local output at the VOCs incinerator site. The local output includes at least: an abnormal event identifier. Warning level, risk score Abnormal operating status identifier and the start timestamp of the evidence capture window. With end timestamp By writing exception event identifiers to the local output. This enables on-site maintenance personnel to identify abnormal events. The encrypted evidence data that uniquely locates the edge monitoring node is located in the protected storage area and corresponds to it. This achieves a closed-loop consistency between early warning prompts and evidence location.

[0040] S620, Early Warning Reporting Packet Encapsulation and Remote Transmission Steps: The edge monitoring node encapsulates the operating condition early warning information into an early warning reporting packet. And send it to the remote monitoring or management platform; early warning reporting package At least include: abnormal event identifiers Warning level, risk score The start timestamp of the evidence capture window With end timestamp Summary of Evidence Package And the range of sampling sequence numbers used for remote deduplication and unique identification. ; through early warning reporting package The fields are fixed, enabling the remote receiver to base its operation on exception event identifiers. Summary of evidence package The warning information is uniquely indexed and the corresponding evidence is located.

[0041] S630, hierarchical reporting, acknowledgment confirmation, and network outage retransmission protection steps: Remote reporting adopts a hierarchical reporting mechanism: edge monitoring nodes send early warning reporting packets first. and receive confirmation receipt from the remote receiver. Then, send the abnormal event identifier according to the preset strategy. One-to-one encrypted evidence data Or its fragmented data; among which, receipt confirmation At least the following conditions must be met: in, For the preset hash function, This indicates a concatenation operation performed in a preset order. This is the timestamp recorded by the remote receiver; if no confirmation is received within the preset timeout period, the network will be interrupted. At that time, the edge monitoring node will report the warning packet. and the corresponding encrypted evidence data Write it into the pending transmission queue and retransmit it according to the retransmission policy until an acknowledgment is received. This is to ensure the reliable delivery and non-loss of early warning information and evidence data.

[0042] S640, Remote Consistency Recalculation Verification and Traceability Review Closed-Loop Steps: The remote receiver receives the warning reporting packet... Then, based on the abnormal event identifier Summary of evidence package Establish a one-to-one correspondence between early warning and evidence; upon receiving an abnormal event identifier... Corresponding encrypted evidence data And decrypt the data packet containing evidence of abnormal operating conditions. Subsequently, the remote receiving end recalculates the evidence packet digest. The summary of the recalculated evidence package satisfies: Summary of the recalculation evidence package Summary of evidence package Compare; when Timely determination of abnormal operating condition evidence data packets The data packet containing evidence of abnormal operating conditions was not tampered with; based on this, the remote receiving end further analyzed the data packet containing evidence of abnormal operating conditions. The hash digest and chain digest of key operational data frames within the evidence chain fragment are recalculated to complete consistency verification, thereby realizing a closed loop of traceable review of operational condition early warning information and its corresponding abnormal operational condition evidence data.

[0043] Example 2: Figure 2 As shown, this embodiment provides an intelligent monitoring and early warning system for the operating conditions of a VOCs incinerator, including: Edge monitoring nodes are deployed at the VOCs incinerator site to establish data acquisition connections with the incinerator control system and field sensors. Multiple operational data acquisition channels are configured to collect real-time data on incinerator combustion chamber temperature, system pressure, inlet and outlet VOCs concentrations, operational status of key actuators, and bypass valve status. Edge monitoring nodes also perform time-aligned processing on the collected operational data according to a unified sampling period, generating a time-ordered raw operational data stream with an appended unified timestamp. The critical operation data processing module is located in the edge monitoring node. It is used to filter key operation data that characterize the effectiveness of VOCs incinerator treatment and operation compliance from the time-ordered raw operation data stream, and encapsulate the key operation data into key operation data frames. The critical operation data processing module is also used to calculate purification efficiency data based on inlet and outlet VOCs concentration data, and write the purification efficiency data as derived data items into the corresponding key operation data frames. The data integrity identifier generation module is set in the edge monitoring node. It is used to generate data integrity identifier information that corresponds to each key running data frame and to bind the key running data frame with the data integrity identifier information for output. The tamper-proof evidence chain construction module is set in the edge monitoring node. It is used to perform sequential consistency verification on key running data frames at consecutive times based on the timestamp field and sequence number field of key running data frames, and generate a chain digest based on the data integrity identification information at the current time and the chain digest at the previous time, thereby constructing an tamper-proof evidence chain formed by connecting multiple evidence chain nodes in chronological order. The consistency assessment and anomaly identification module is set up in the edge monitoring node. It is used to construct a feature vector reflecting the consistency of the operational mechanism and calculate the consistency deviation, provided that the anti-tampering evidence chain consistency verification passes. The module also introduces an ant colony algorithm onto a preset operational mode graph to perform multi-path consistency assessment on the consistency deviation within a sliding time window, obtain the optimal consistency path, and calculate the consistency matching degree based on the optimal consistency path. Furthermore, the module introduces a reinforcement learning mechanism to adaptively update the consistency assessment parameters and identify abnormal operational states based on the consistency matching degree and preset judgment rules. The abnormal operating condition evidence solidification and early warning generation module is set in the edge monitoring node. When an abnormal operating state is detected, it is used to extract evidence chain fragments containing data before and after the occurrence of the abnormal operating state from the anti-tamper evidence chain, and encapsulate the evidence chain fragments into an abnormal operating condition evidence data package. The abnormal operating condition evidence solidification and early warning generation module is also used to perform digest calculation and encryption encapsulation on the abnormal operating condition evidence data package, and calculate risk scores based on the persistence, frequency of occurrence, consistency deviation and reinforcement learning output results of the abnormal operating state, and then generate operating condition early warning information with corresponding warning levels. The early warning output and remote reporting module is set in the edge monitoring node. It is used to output local early warning information at the VOCs incinerator site, including abnormal event identifier, early warning level, risk score and evidence time range. The early warning output and remote reporting module is also used to encapsulate the operating condition early warning information into an early warning reporting package and send it to the remote monitoring or management platform. After receiving the acknowledgment, it reports the encrypted abnormal operating condition evidence data corresponding one-to-one with the abnormal event identifier. The remote consistency verification module is set up in the remote monitoring or management platform. It is used to establish a one-to-one correspondence between early warning information and abnormal operating condition evidence data based on abnormal event identifiers and evidence package summaries. It also performs summary recalculation and evidence chain consistency verification on the received abnormal operating condition evidence data to achieve traceable review of operating condition early warning information and its corresponding abnormal operating condition evidence data.

[0044] This invention unifies the collection, time alignment, and consistency analysis of key operating condition data such as combustion chamber temperature, inlet and outlet VOCs concentrations, bypass valve status, and actuator operation status of VOCs incinerators. Based on tamper-proof evidence chains, ant colony algorithms, and reinforcement learning, it achieves intelligent identification and early warning of operating status. This invention can ensure that pollutants are removed in compliance with standards while constraining the incinerator to always operate within the operating range that meets the conditions for efficient combustion in real time. It effectively avoids energy waste and secondary pollution caused by insufficient or excessive combustion, thereby improving the stability and regulatory capacity of VOCs treatment and achieving synergistic optimization of air pollution control and efficient combustion technology.

[0045] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0046] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0047] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0048] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0049] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0050] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0051] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0052] If a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0053] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0054] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent monitoring and early warning of the operating conditions of a VOCs incinerator, characterized in that, Includes the following steps: S1. Edge monitoring nodes collect operating condition data from the VOCs incinerator's operating system in real time and attach a unified timestamp to the collected operating condition data to form a time-ordered raw operating condition data stream. S2. Select key operational data to reflect the effectiveness of VOCs incinerator pollution control and operational compliance, and generate data integrity identification information for the key operational data in chronological order, so that the key operational data at each moment and its corresponding data integrity identification information form a one-to-one correspondence. S3. Construct a continuous and tamper-proof evidence chain in chronological order, so that the key operational data at the current moment and the key operational data at the previous moment form an irreversible logical relationship, thereby achieving tamper-proof protection of the continuity of key operational data. S4. Utilize the tamper-proof evidence chain to verify the consistency of key operational data. By analyzing the logical consistency between inlet and outlet VOCs concentration data, bypass valve status data, and operational status data of key actuators, identify abnormal operational states that are inconsistent with the normal operating mechanism. S5. When an abnormal operating state is detected, the abnormal operating condition evidence solidification process is triggered. The tamper-proof evidence chain fragment corresponding to the abnormal operating state is encrypted, encapsulated, and stored as abnormal operating condition evidence data. At the same time, the corresponding level of operating condition warning information is generated according to the degree of abnormality.

2. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 1, characterized in that, It also includes S6, which outputs and reports operational condition warning information locally and remotely, and uploads integrity verification information of abnormal operating condition evidence data simultaneously during the reporting process, so that the receiving end can perform consistency verification and traceability review of operational condition warning information and its corresponding abnormal operating condition evidence data.

3. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 1, characterized in that, S1 specifically refers to: By deploying edge monitoring nodes at the VOCs incinerator site, a data acquisition connection is established with the incinerator control system and on-site sensors to obtain operating condition data that reflects the incinerator's operating status. According to a unified sampling cycle, real-time data on combustion chamber temperature, system pressure, inlet and outlet VOCs concentration, operating status of key actuators, and bypass valve status are collected from the operating condition data acquisition channel to form raw operating condition data. Assign timestamp information under a unified time base to each piece of original operating condition data to ensure that the original operating condition data has sequential consistency in the time dimension; The raw operating condition data with added timestamps are arranged in the order of collection time to form a time-ordered raw operating condition data stream, which is then used as input data for subsequent key operating data filtering.

4. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 1, characterized in that, S2 specifically refers to: Based on the pollution control mechanism of VOCs incinerators, key operational data for characterizing the compliance and effectiveness of incinerator operation were selected from the original operating condition data stream; Key operational data should include at least the inlet and outlet VOCs concentration data directly related to pollutant removal, the bypass valve status data directly related to the operating status of the treatment facility, and the combustion chamber temperature data related to the combustion process.

5. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 4, characterized in that, S2 also includes: For each key operational data obtained through screening, data integrity identification information is generated according to its corresponding timestamp, so that a one-to-one correspondence is formed between key operational data and data integrity identification information. The key operational data that forms the corresponding relationship and its data integrity identification information are output as the basic data units for constructing an anti-tampering evidence chain.

6. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 1, characterized in that, S3 specifically refers to: The basic data units are arranged sequentially according to the timestamp order to determine the temporal relationship between key operational data at adjacent times. The key operational data and its data integrity identification information at the current moment are correlated and calculated with the integrity identification information of the key operational data at the previous moment to generate a continuous correlation relationship; Based on continuous correlation, an anti-tampering evidence chain covering continuous time periods is constructed, so that key operational data form an irreversible time sequence correlation in logical structure; The completed tamper-proof evidence chain will serve as the trusted data foundation for subsequent consistency verification and anomaly identification.

7. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 1, characterized in that, S4 specifically refers to: Based on the tamper-proof evidence chain, the consistency of key operational data within a continuous time period is verified, and the logical relationship between inlet and outlet VOCs concentration data, bypass valve status data, and key actuator operational status data is analyzed. During the consistency verification process, an ant colony algorithm is introduced to construct a multi-path consistency evaluation model. By simulating various possible normal operation paths, the consistency matching degree of the current key operating data under different paths is comprehensively evaluated. By combining reinforcement learning mechanisms, the parameters of the consistency evaluation model are adaptively updated, enabling the model to continuously strengthen its ability to identify normal operating modes based on historical operating results. When the matching degree of the current key operating data in the consistency assessment model is consistently lower than the preset consistency threshold, the corresponding operating status will be identified as an abnormal operating status, and an abnormal operating status identifier will be output.

8. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 1, characterized in that, S5 specifically refers to: After identifying an abnormal operating state, the abnormal operating condition evidence solidification process is triggered, and the evidence chain segment corresponding to the abnormal operating state is extracted from the tamper-proof evidence chain. The evidence chain fragments are encrypted and encapsulated to form abnormal working condition evidence data, and the abnormal working condition evidence data is stored in the protected storage area of ​​the edge monitoring node; Based on the duration, frequency, and degree of consistency deviation of abnormal operating states, and combined with the risk assessment results obtained from reinforcement learning, the warning level of abnormal operating states is determined. Generate operating condition warning information corresponding to the warning level, and associate and mark the operating condition warning information with abnormal operating condition evidence data.

9. The intelligent monitoring and early warning method for the operating conditions of a VOCs incinerator according to claim 2, characterized in that, S6 specifically refers to: The operating condition warning information will be output locally in a visual manner at the VOCs incinerator site to alert on-site maintenance personnel to the current abnormal operating status. The operational status warning information is sent to the remote monitoring or management platform via remote communication. While sending operational condition warning information, the system simultaneously uploads integrity verification information of the associated abnormal operating condition evidence data to support remote verification of the authenticity of the abnormal operating condition evidence data. This enables remote receivers to perform consistency verification and traceability review of operational condition warning information and its corresponding abnormal operating condition evidence data based on integrity verification information, forming a complete closed loop of warning and evidence collection.

10. A VOCs incinerator operating condition intelligent monitoring and early warning system, based on the VOCs incinerator operating condition intelligent monitoring and early warning method according to any one of claims 1-9, characterized in that, include: Edge monitoring nodes are deployed at the VOCs incinerator site to establish data acquisition connections with the incinerator control system and on-site sensors, and are configured with multiple operating condition data acquisition channels. The key operational data processing module, located in the edge monitoring node, is used to filter key operational data that characterizes the effectiveness of VOCs incinerator treatment and operational compliance from the time-ordered raw operational data stream; The data integrity identifier generation module is set in the edge monitoring node and is used to generate data integrity identifier information that corresponds to each key running data frame. The tamper-proof evidence chain construction module is set in the edge monitoring node and is used to perform sequential consistency verification on key operational data frames at consecutive times based on the timestamp field and sequence number field of key operational data frames. The consistency assessment and anomaly identification module is set up in the edge monitoring node. It is used to construct a feature vector that reflects the consistency of the operation mechanism, provided that the consistency verification of the anti-tampering evidence chain is passed. The abnormal operating condition evidence solidification and early warning generation module is set in the edge monitoring node. It is used to extract evidence chain fragments containing data before and after the occurrence of the abnormal operating state from the anti-tamper evidence chain when an abnormal operating state is identified. The early warning output and remote reporting module is set in the edge monitoring node and is used to output local early warning information at the VOCs incinerator site, including abnormal event identifier, early warning level, risk score and evidence time range; The remote consistency verification module is set up in a remote monitoring or management platform to establish a one-to-one correspondence between early warning information and abnormal operating condition evidence data based on abnormal event identifiers and evidence package summaries.

Images