A protocol conversion method of an internet of things water information heterogeneous data transmission protocol gateway

Abstract

The application relates to the technical field of Internet of Things, in particular to a protocol conversion method of an Internet of Things water information heterogeneous data transmission protocol gateway, which comprises the following steps: constructing a protocol feature time sequence atlas, generating a protocol difference matrix based on the protocol feature time sequence atlas; extracting cross-protocol conversion features based on the protocol difference matrix, constructing a heterogeneous protocol conversion model, and obtaining a protocol conversion feature vector; mapping the protocol conversion feature vector to a rule space to obtain an initial protocol conversion rule set; adjusting the initial protocol conversion rule set to obtain a target protocol conversion rule set; and converting the target protocol conversion rule set into a transmission instruction recognizable by equipment and executing the same. Through real-time tracking and optimization of protocol conversion accuracy, transmission time delay and data integrity, any potential problem in the protocol conversion process can be dynamically adjusted, and accurate data transmission and timely feedback are ensured.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, specifically to a protocol conversion method for a gateway for heterogeneous data transmission protocols of water information in the Internet of Things. Background Technology

[0002] In the field of water resource management, IoT devices can include various types of sensors, such as water level sensors, flow sensors, water quality sensors, and meteorological sensors. They monitor various physical, chemical, and biological properties of water in different locations and environments. The data generated by these devices often uses different protocols, data formats, sampling frequencies, and even different communication methods (such as LoRa, NB-IoT, Wi-Fi, etc.), thus forming "heterogeneous data".

[0003] Currently, traditional protocol conversion usually relies on fixed mapping relationships or manual configuration methods, lacking in-depth analysis of protocol characteristics and time-series correlation modeling. It cannot accurately identify subtle differences between different protocols, resulting in poor compatibility between protocols and data loss or field misalignment during the conversion process.

[0004] Furthermore, existing technologies often use static rule sets for protocol conversion, which cannot monitor and dynamically adjust transmission delay, data integrity, or conversion accuracy in real time. Once the network environment or data type changes, the system is difficult to respond in time, which can easily lead to increased transmission delay or accumulation of data errors. Summary of the Invention

[0005] To achieve the above objectives, the present invention provides the following technical solution: a protocol conversion method for an IoT water information heterogeneous data transmission protocol gateway, comprising:

[0006] Obtain heterogeneous protocol characteristics and historical transmission data of IoT water information devices, construct a protocol feature time series map, and generate a protocol difference matrix based on the protocol feature time series map;

[0007] Based on the protocol difference matrix, cross-protocol conversion features are extracted, a heterogeneous protocol conversion model is constructed, and a protocol conversion feature vector is obtained.

[0008] The protocol conversion feature vectors are mapped to the rule space to obtain the initial protocol conversion rule set;

[0009] The initial protocol conversion rule set is adjusted to obtain the target protocol conversion rule set;

[0010] The target protocol conversion rule set is converted into a device-recognizable transmission command and executed.

[0011] Dynamically track protocol conversion accuracy, transmission latency, and data integrity to obtain optimized transmission instructions.

[0012] Preferably, constructing a temporal map of protocol features includes:

[0013] The heterogeneous protocol features are matched and associated with historical transmission data to obtain effective association pairs; based on the correspondence between protocol attribute categories and data attribute categories, groups are formed to obtain datasets with the same attribute association.

[0014] Based on the effective association pairs in the same attribute association dataset, the core attributes and temporal attributes of the graph nodes are defined; based on the association strength between the heterogeneous protocol features and historical transmission data, the association weights and temporal constraints of the graph edges are defined; based on the node attributes, edge association weights, and temporal constraints, the basic framework of the temporal graph is constructed; the temporal association paths of transmission data between different nodes are established, and the effective association pairs are bound to the nodes and edges in temporal order to obtain the initial protocol feature temporal graph;

[0015] Real-time transmission data from the IoT water information device is collected. Based on the node association and data transmission efficiency of the initial protocol feature time-series graph, a graph adaptation coefficient is calculated. The graph adaptation coefficient is compared with a preset adaptation threshold to obtain an adaptation comparison result. Based on the adaptation comparison result, the current adaptation state of the initial protocol feature time-series graph is determined. If the adaptation state does not meet the preset requirements, the core attributes and temporal attributes of the nodes in the initial protocol feature time-series graph are updated. The association weights and temporal constraints of the edges in the initial protocol feature time-series graph are adjusted to obtain the protocol feature time-series graph.

[0016] Preferably, generating a protocol difference matrix based on the protocol feature time-series graph includes:

[0017] The node attributes and edge association information of the temporal graph of protocol features are analyzed, the static feature vector of the protocol and the temporal dynamic association strength are extracted, the feature comparison dimension and the association comparison dimension are determined, and the feature benchmark set and the association benchmark set are obtained.

[0018] Establish the deviation quantification rule for the feature comparison dimension and the variation measurement rule for the association comparison dimension; perform item-by-item comparison of the protocol static feature vector based on the feature benchmark set to obtain the feature deviation degree; perform stage comparison of the time-series dynamic association strength based on the association benchmark set to obtain the association variation degree, and group according to the comparison dimension to obtain the same dimension difference dataset.

[0019] A matrix framework is constructed with the feature comparison dimension as the matrix row dimension and the association comparison dimension as the matrix column dimension. The feature deviation and association variability in the same dimension difference dataset are mapped to the corresponding positions in the matrix framework. The difference level classification standard is determined, the difference status of each position element is labeled, and the initial protocol difference matrix is ​​obtained.

[0020] Collect real-time protocol interaction data, calculate adjustment coefficients based on the element distribution characteristics of the initial protocol difference matrix, update the values ​​of feature deviation and correlation variability based on the adjustment coefficients, and synchronously update the difference status labels of elements at each position to obtain the protocol difference matrix.

[0021] Preferably, cross-protocol conversion features are extracted based on the protocol difference matrix to construct a heterogeneous protocol conversion model, resulting in a protocol conversion feature vector, including:

[0022] The frequency of occurrence of each difference attribute in different protocol pairs in the protocol difference matrix is ​​statistically analyzed, and high-frequency difference attributes are selected as basic features. The co-occurrence relationship between the basic features is analyzed to obtain associated features. The basic features and the associated features are hierarchically divided, and the hierarchical level of each feature and the dependency relationship between the levels are marked. The basic features, the associated features and the hierarchical dependency relationship are integrated to obtain a cross-protocol conversion feature set.

[0023] A heterogeneous protocol conversion model is constructed based on the cross-protocol conversion feature set. The cross-protocol conversion feature set is used as the feature source of the model input layer. A feature mapping module and a conversion rule generation module are set up. The feature mapping module is used for feature dimension adaptation, and the conversion rule generation module is used to generate conversion logic based on the adapted features.

[0024] The feature mapping module receives the cross-protocol conversion feature set, performs dimensional alignment based on the hierarchical attributes of each feature, and maps features at different levels to a unified dimensional space to obtain normalized features; the conversion rule generation module assigns weights to the normalized features; and generates dynamic conversion rules based on the weight assignment results and the normalized features, wherein the dynamic conversion rules include conversion priorities and mapping relationships between features; the feature parameters of the dynamic conversion rules are fused with the weight values ​​of the normalized features to obtain a protocol conversion feature vector.

[0025] Preferably, the protocol conversion feature vector is mapped to a rule space to obtain an initial protocol conversion rule set, including:

[0026] The protocol transformation feature vector is mapped to the initial rule space to obtain the initial rule vector;

[0027] Based on the water information data transmission standard, the effective rule probability of the initial rule vector is obtained;

[0028] Based on the effective rule probabilities, and based on the adaptive rule pruning method and feature space correction, the effective rule probabilities are optimized to obtain the initial protocol conversion rule set.

[0029] Preferably, the initial protocol conversion rule set is adjusted to obtain the target protocol conversion rule set, including:

[0030] The initial protocol conversion rule set is adjusted by multi-objective collaborative optimization, and a subset of efficient rules is extracted through performance screening to obtain a candidate conversion rule set;

[0031] The candidate conversion rule set is converted into field conversion parameters to obtain the target protocol conversion rule set.

[0032] Preferably, converting the target protocol conversion rule set into device-recognizable transmission instructions and executing them includes:

[0033] The target protocol conversion rule set is parsed based on structured protocol parsing to extract device identifiers, field conversion parameters and transmission timing to obtain the original transmission instructions;

[0034] The water information protocol adapter converts the original transmission command into a device-compatible protocol format; based on the transmission encryption mechanism, the transmission command is sent and executed through a secure transmission channel.

[0035] Preferably, the protocol conversion accuracy, transmission latency, and data integrity are dynamically tracked to obtain optimized transmission instructions, including:

[0036] The real-time transmission data stream and corresponding conversion logs during the protocol conversion process are obtained. The real-time transmission data stream and conversion logs are cleaned, deduplicated, and formatted to obtain the basic transmission dataset.

[0037] Based on the aforementioned basic transmission dataset, protocol conversion accuracy features, transmission delay features, and data integrity features are extracted respectively to construct a dynamic tracking model. The dynamic tracking model is used to monitor and record the protocol conversion accuracy features, transmission delay features, and data integrity features in real time to obtain a real-time tracking feature sequence.

[0038] Anomaly identification and correlation analysis are performed on the real-time tracking feature sequence to determine feature anomalies and anomaly correlation relationships, and an anomaly correlation map is obtained;

[0039] Based on the aforementioned anomaly correlation graph and a preset transmission quality benchmark, protocol conversion accuracy constraints, transmission delay constraints, and data integrity constraints are determined to obtain a multi-objective constraint parameter set.

[0040] For each abnormal association in the abnormal association graph, the characteristic combination conditions that trigger the relationship are analyzed; based on the characteristic combination conditions and the multi-objective constraint parameter set, the corresponding transmission parameter adjustment scheme is determined, wherein the transmission parameters include data packet size, retransmission mechanism and encoding method; the transmission parameter adjustment scheme is mapped and stored with the corresponding abnormal association to obtain a strategy library;

[0041] The strategy library is matched and simulated with the real-time tracking feature sequence. The optimal adjustment strategy is determined through iterative optimization to obtain the optimized transmission command.

[0042] Preferably, based on the anomaly correlation graph and a preset transmission quality benchmark, protocol conversion accuracy constraints, transmission delay constraints, and data integrity constraints are determined to obtain a multi-objective constraint parameter set, including:

[0043] The influence weights of each characteristic anomaly type are extracted from the anomaly correlation graph, wherein the influence weights represent the degree of interference of the anomaly type on the overall transmission quality;

[0044] Based on the aforementioned influence weights, the preset transmission quality benchmark is dynamically adjusted to determine the maximum allowable deviation in protocol conversion accuracy, the longest transmission delay threshold, and the minimum standard for data integrity.

[0045] The maximum deviation of the protocol conversion accuracy, the longest transmission delay threshold, and the minimum standard of data integrity are integrated to obtain a multi-objective constraint parameter set.

[0046] Compared with the prior art, the beneficial effects of the present invention are:

[0047] This invention, by constructing a protocol feature time series map and generating a protocol difference matrix, can accurately identify the differences between different protocols, and then extract cross-protocol conversion features. This can ensure data interoperability and compatibility between heterogeneous protocol devices, improve the overall efficiency of the system, and reduce protocol mismatch and data loss.

[0048] This invention, through real-time tracking and optimization of protocol conversion accuracy, transmission latency, and data integrity, can dynamically adjust for any potential problems during the protocol conversion process, ensuring accurate data transmission and timely feedback; real-time monitoring and anomaly correlation analysis can help identify and repair problems in transmission in a timely manner, thereby optimizing network resource utilization and improving the reliability of data transmission.

[0049] This invention ensures data security by employing a transmission encryption mechanism during the transmission command conversion process. Simultaneously, it further enhances the stability and security of data transmission by optimizing and adjusting data packet size, retransmission mechanisms, and encoding methods based on anomaly correlation graphs, thus ensuring data integrity. Furthermore, by comparing graph adaptation coefficients with thresholds, the protocol feature graph can be adaptively adjusted to optimize the data transmission adaptation state, ensuring high efficiency in real-time data transmission, reducing adaptation errors during transmission, and thereby optimizing the accuracy of the protocol conversion process.

[0050] This invention utilizes multi-objective collaborative optimization technology to adjust the protocol conversion rule set, which can dynamically select the optimal conversion rule according to actual needs, avoid redundant operations and reduce computational complexity, so that not only is efficiency maintained during transmission, but flexibility is also improved, adapting to different application scenarios and changing needs. Attached Figure Description

[0051] Figure 1 This is a schematic flowchart of the overall method in one embodiment of the present invention. Detailed Implementation

[0052] 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.

[0053] Example 1, please refer to Figure 1 This invention provides a technical solution: a protocol conversion method for an IoT water information heterogeneous data transmission protocol gateway, comprising:

[0054] S1. Obtain the heterogeneous protocol characteristics and historical transmission data of IoT water information devices, construct a protocol feature time series map, and generate a protocol difference matrix based on the protocol feature time series map;

[0055] S2. Extract cross-protocol conversion features based on the protocol difference matrix, construct a heterogeneous protocol conversion model, and obtain the protocol conversion feature vector;

[0056] S3. Map the protocol conversion feature vectors to the rule space to obtain the initial protocol conversion rule set;

[0057] S4. Adjust the initial protocol conversion rule set to obtain the target protocol conversion rule set;

[0058] S5. Convert the target protocol conversion rule set into a transmission command that the device can recognize and execute it;

[0059] S6. Dynamically track protocol conversion accuracy, transmission delay, and data integrity to obtain optimized transmission instructions.

[0060] It should be noted that different brands and types of water equipment may use different communication protocols. For example, some equipment may use the 206 protocol, while others may use the 651 protocol. The 206 protocol typically refers to a communication protocol used in water and environmental monitoring, enabling data transmission and exchange between different devices and systems. This protocol is mainly used in the industry to standardize equipment communication for remote monitoring, data acquisition, and control. The 651 protocol, on the other hand, is a communication protocol used in water and environmental monitoring, primarily for data transmission and control between devices. In practical applications, the 651 protocol enables efficient data exchange, allowing equipment from different manufacturers to communicate. These protocols have their own communication methods, data formats, and other characteristics, and are therefore incompatible with each other.

[0061] This process involves collecting characteristics of the protocols used by these devices when transmitting data, such as data frame format and data transmission methods. Analyzing historical data transmissions helps identify the device's behavior patterns and data transmission characteristics under different protocols. The characteristics of different protocols are then arranged chronologically to form a timeline. Here, "timeline" refers to the changes in the device's protocol usage patterns over a period of time. For example, some protocols may transmit large amounts of data over a period, while others may transmit data periodically. Based on these timeline maps, a matrix can be generated to represent the differences between different protocols. These differences may include variations in data format, transmission rate, error handling mechanisms, and other aspects.

[0062] The cross-protocol conversion features are extracted from the protocol difference matrix; for example, how to map data structures in one protocol to data structures in another protocol. Based on these conversion features, a model is built for data conversion between different protocols. This model can understand the differences between different protocols and make corresponding conversions. By using a heterogeneous protocol conversion model, the protocol conversion features can be represented as a "feature vector." This vector is the mathematical representation of the protocol conversion and can help the system understand how to convert from one protocol to another. The rule space refers to the set of all possible protocol conversion rules. This space defines how to convert data from protocol A to the format of protocol B. Mapping the feature vector of protocol conversion to the rule space yields a preliminary set of protocol conversion rules. These rules may be determined based on factors such as device type and data content. The initial set of protocol conversion rules may not be perfect, so adjustments and optimizations are needed. For example, some rules may not work in practical applications, or may cause data loss or transmission delays in certain special cases. After adjustments, the final set of protocol conversion rules is formed, which can complete the protocol conversion task more efficiently and accurately.

[0063] The target protocol conversion rule set is converted into specific device instructions, which the device can recognize and execute. These instructions specify how to use the target protocol for data transmission and communication. After executing these instructions, the device will be able to communicate and exchange data between different protocols. By monitoring the accuracy of protocol conversion, transmission latency, and data integrity in real time, the system can understand the performance during the protocol conversion process. Based on real-time data feedback, the system can optimize transmission instructions. For example, if it finds that certain conversion rules are causing data loss or excessive latency, the system can dynamically adjust the rules to ensure data integrity and transmission efficiency.

[0064] In an optional embodiment, constructing a temporal graph of protocol features includes:

[0065] The heterogeneous protocol features are matched and associated with historical transmission data to obtain effective association pairs; the data are grouped based on the correspondence between protocol attribute categories and data attribute categories to obtain datasets with the same attribute association.

[0066] Based on the effective association pairs in the same attribute association dataset, the core attributes and temporal attributes of the graph nodes are defined; based on the association strength between heterogeneous protocol features and historical transmission data, the association weights and temporal constraints of the graph edges are defined; based on the node attributes, edge association weights, and temporal constraints, the basic framework of the temporal graph is constructed; the temporal association paths of transmission data between different nodes are established, and the effective association pairs are bound to the nodes and edges in temporal order to obtain the initial protocol feature temporal graph;

[0067] Real-time transmission data from IoT water information devices is collected. Based on the node association and data transmission efficiency of the initial protocol feature time-series graph, the graph adaptation coefficient is calculated. The graph adaptation coefficient is compared with a preset adaptation threshold to obtain the adaptation comparison result. Based on the adaptation comparison result, the current adaptation status of the initial protocol feature time-series graph is determined. If the adaptation status does not meet the preset requirements, the core attributes and temporal attributes of the nodes in the initial protocol feature time-series graph are updated. The association weights and temporal constraints of the edges in the initial protocol feature time-series graph are adjusted to obtain the protocol feature time-series graph.

[0068] It should be noted that by comparing the characteristics of the protocol with historical transmission data, the correspondence between protocol characteristics and actual transmission data is found; for example, how a specific data frame format of a certain protocol is represented in actual data; the final "effective correlation pair" refers to a data pair that has a clear correlation between protocol characteristics and historical data and can reflect the protocol behavior; protocol characteristics can be classified according to different dimensions, such as protocol transmission method, data format, latency requirements, etc.; data characteristics can also be classified, such as data size, transmission frequency, precision requirements, etc.; according to the correspondence between protocol attributes and data attributes, the correlation pairs in historical transmission data are grouped according to attribute categories; the purpose of this is to ensure that data with similar attributes can be aggregated together during protocol conversion, reducing unnecessary conversion errors;

[0069] In the graph, each node represents a protocol feature or data attribute. The core attributes of these nodes refer to the basic characteristics of the protocol itself, such as data format and communication rate. Protocol behavior may change over time; therefore, temporal attributes refer to the protocol's performance at different times. For example, some protocols may experience peak data transmission during specific time periods, while other periods are relatively quiet. Based on valid association pairs in the same attribute-related dataset, each graph node is assigned its core attribute (protocol feature) and temporal attribute (time-varying characteristics). The correlation strength between protocol features and historical data indicates the closeness of their relationship; a higher correlation strength means a stronger match between the protocol feature and the data. In the graph, edges connect different nodes, representing relationships between protocols or data flows. Each edge is assigned a weight based on its correlation strength; a larger weight indicates a higher level of trust in the transition between protocols. Temporal constraints refer to the temporal constraints between protocols. For example, some protocols may have time-sequence requirements for transmission or require data exchange to be completed within a specific timeframe. Temporal constraints ensure the rationality of the temporal graph.

[0070] By combining the core attributes of nodes, the association weights of edges, and temporal constraints, a basic framework for a temporal graph is constructed. This framework defines the relationships between nodes (protocol features or data attributes) and how data transmission and protocol conversion occur over time. The temporal association path of transmitted data refers to the time sequence path of data transmission between protocols. Through temporal constraints, the transmission order of data from one protocol to another is defined. Valid association pairs are bound to nodes (protocols) and edges (relationships) in the graph in chronological order to construct a preliminary protocol feature temporal graph. During the operation of IoT devices, their actual real-time transmission data is collected. The graph adaptation coefficient is a coefficient that measures the degree of matching between the initial protocol feature temporal graph and the real-time transmission data; the higher the adaptation coefficient, the better the graph adapts to actual transmission needs. Based on the association between real-time data and graph nodes, the adaptation coefficient is calculated to evaluate the effectiveness of the protocol feature temporal graph.

[0071] The preset adaptation threshold is a predefined adaptation standard value, representing the minimum matching degree required during protocol conversion. By comparing the calculated graph adaptation coefficient with the preset threshold, it is determined whether the current graph adaptation status meets the requirements. If the graph adaptation coefficient is lower than the preset threshold, it indicates that the protocol adaptation effect is not ideal, and the graph needs to be adjusted. The nodes (protocol features) in the graph are updated, and their core attributes and temporal attributes are adjusted to adapt to the actual data transmission requirements. At the same time, it is also necessary to adjust the association weights and temporal constraints of the edges between nodes to optimize the protocol conversion process, and finally obtain a more optimized protocol feature temporal graph.

[0072] In an optional embodiment, generating a protocol difference matrix based on a protocol feature time series graph includes:

[0073] The node attributes and edge association information of the temporal graph of protocol features are analyzed, the static feature vector of the protocol and the temporal dynamic association strength are extracted, the feature comparison dimension and the association comparison dimension are determined, and the feature benchmark set and the association benchmark set are obtained.

[0074] Establish deviation quantification rules for feature comparison dimension and variation measurement rules for association comparison dimension; perform item-by-item comparison of protocol static feature vectors based on feature benchmark set to obtain feature deviation degree; perform stage comparison of temporal dynamic association strength based on association benchmark set to obtain association variation degree, and group according to comparison dimension to obtain the same dimension difference dataset.

[0075] A matrix framework is constructed with the feature comparison dimension as the matrix row dimension and the association comparison dimension as the matrix column dimension. The feature deviation and association variability in the difference dataset of the same dimension are mapped to the corresponding positions in the matrix framework. The difference level classification standard is determined, the difference status of each element is labeled, and the initial protocol difference matrix is ​​obtained.

[0076] Real-time protocol interaction data is collected, adjustment coefficients are calculated based on the element distribution characteristics of the initial protocol difference matrix, the values ​​of feature deviation and correlation variability are updated based on the adjustment coefficients, and the difference status labels of elements at each position are updated synchronously to obtain the protocol difference matrix.

[0077] It should be noted that the protocol feature time-series graph consists of protocol nodes and edges between them. Nodes represent protocol features (e.g., protocol type, version, data transmission method, etc.), and edges represent the relationships between these features (e.g., the transmission order or data dependency of different protocol features). The attributes of nodes are the static features of the protocol, while the association information of edges describes the temporal relationships and data flow patterns between protocol features. The static features of the protocol include attributes that do not change over time, such as the protocol's data format, transmission rate, signal strength, etc. By analyzing the attributes of the nodes, the static feature vector of each protocol is extracted. The temporal dynamic correlation strength refers to the degree of correlation between protocol features over time, mainly describing the fluctuation of the relationship between nodes over time. For example, some protocol features may have a strong correlation at a specific time, while the correlation is weaker at other times.

[0078] Feature comparison dimensions represent the dimensions that can be compared within the static feature vectors of the protocols; for example, the transmission rate and data format of the protocols can be used as comparison dimensions. Correlation comparison dimensions represent the dimensions that can be compared within the temporal dynamic correlation strength, typically involving the correlation strength between different protocols at different points in time. The feature benchmark set and correlation benchmark set respectively contain elements from the static feature vectors and temporal dynamic correlation strength that can serve as comparison benchmarks. Deviation quantification rules are used to measure the deviation between the static feature vectors of the protocols; possible quantification methods include calculating the differences and variances of each dimension. Variation measurement rules are used to measure the variation in temporal dynamic correlation strength, typically by comparing the correlation strength at different points in time to determine the magnitude of the change. Each static feature vector in the feature benchmark set is compared with the actual static features of the protocol, and the deviation is calculated one by one; this typically involves measurements of numerical differences, trends, etc.

[0079] The temporal dynamic correlation strength is compared across different time periods, and the variability of each stage is calculated to assess the temporal correlation changes between protocols. The calculated feature deviation and correlation variability are grouped according to different comparison dimensions to obtain a difference dataset for each dimension; for example, transmission rate deviation and temporal correlation strength variation can be used as separate datasets. A two-dimensional matrix is ​​constructed, where rows represent feature comparison dimensions and columns represent correlation comparison dimensions. The position of each element in the matrix represents the degree of difference in a specific feature and correlation dimension. The feature deviation and correlation variability of each dimension are mapped to the matrix according to their corresponding positions within the matrix framework; each element represents the degree of difference in a specific dimension. Based on the values ​​of deviation and variability, difference levels are defined, such as low difference, medium difference, and high difference, for subsequent analysis. According to the preset difference level standards, the elements at each position in the matrix are labeled as different difference states (e.g., normal, warning, abnormal, etc.), thus obtaining a preliminary protocol difference matrix.

[0080] The system collects real-time protocol data from the device during actual operation and analyzes the behavior of the current protocol using this data. Based on the distribution characteristics of the elements in the protocol difference matrix, an adjustment coefficient is calculated to update the deviation and variability. Based on the adjustment coefficient, the characteristic deviation and variability of the protocol's temporal correlation strength are updated. Based on the updated deviation and variability, the difference status of each position in the matrix is ​​re-labeled. Finally, the updated matrix contains the protocol difference information adjusted based on real-time data. This matrix is ​​used to evaluate and optimize the adaptability of the protocol, helping to promptly identify and resolve differences and adaptation issues between protocols.

[0081] In an optional embodiment, cross-protocol conversion features are extracted based on the protocol difference matrix to construct a heterogeneous protocol conversion model, resulting in a protocol conversion feature vector, including:

[0082] The frequency of occurrence of each difference attribute in different protocol pairs in the statistical protocol difference matrix is ​​counted, and high-frequency difference attributes are selected as basic features. The co-occurrence relationship between basic features is analyzed to obtain associated features. The basic features and associated features are hierarchically divided, and the hierarchy to which each feature belongs and the dependency relationship between the hierarchy are marked. The basic features, associated features and hierarchical dependencies are integrated to obtain a cross-protocol conversion feature set.

[0083] A heterogeneous protocol conversion model is constructed based on a cross-protocol conversion feature set. The cross-protocol conversion feature set is used as the feature source of the model input layer. A feature mapping module and a conversion rule generation module are set up. The feature mapping module is used for feature dimension adaptation, and the conversion rule generation module is used to generate conversion logic based on the adapted features.

[0084] The feature mapping module receives a cross-protocol conversion feature set, performs dimensional alignment based on the hierarchical attributes of each feature, and maps features at different levels to a unified dimensional space to obtain normalized features. The conversion rule generation module assigns weights to the normalized features. Based on the weight assignment results and the normalized features, dynamic conversion rules are generated, whereby the dynamic conversion rules include the conversion priority and mapping relationship between features. The feature parameters of the dynamic conversion rules are fused with the weight values ​​of the normalized features to obtain the protocol conversion feature vector.

[0085] It should be noted that by statistically analyzing the frequency of each difference attribute in different protocol pairs, we can identify which features appear frequently and which differences are universal. These frequently occurring difference attributes are considered "fundamental features" because they reflect the core differences between protocols and are usually crucial for protocol conversion. Fundamental features may not exist independently but are related to each other; for example, some protocol features may always change together, or there may be significant correlations between certain features. By analyzing the co-occurrence of fundamental features, we can obtain associated features, that is, the interactions or dependencies between these features; for example, "transmission rate" and "latency" may be highly correlated in some protocols, belonging to the same set of associated features. Based on the importance, dependency, and interrelationship of features, fundamental features and associated features are hierarchically classified; for example, some fundamental features may be direct driving factors for protocol conversion (such as transport protocol type), while other associated features may be derived from them (such as error handling methods, data packetization methods, etc.).

[0086] Features at different levels may have dependencies; for example, higher-level protocol features may depend on lower-level protocol features. During protocol conversion, these dependencies must be understood to ensure data consistency. The cross-protocol conversion feature set contains all features used to describe protocol differences, including not only basic and related features but also their hierarchical relationships. The cross-protocol conversion feature set represents all the key information that may affect protocol conversion, combining static and dynamic features of the protocol. The goal of the heterogeneous protocol conversion model is to achieve conversion between different protocols. By inputting the protocol's feature set (cross-protocol conversion feature set), the model can identify the differences between the source and target protocols and generate corresponding conversion rules. The feature mapping module is used for feature dimension adaptation, mapping the features of the source protocol to the feature dimensions required by the target protocol. This means adjusting the feature dimensions to ensure alignment during conversion. Different protocols may have different feature dimension structures; the feature mapping module is responsible for resolving this structural mismatch.

[0087] The main function of the transformation rule generation module is to generate transformation rules based on features mapped to a unified dimension. These rules define how to transform from the source protocol to the target protocol, including the mapping relationship between features, how the data is reformatted, and feature adjustments. In cross-protocol transformations, the feature dimensions of the source protocol and the target protocol may differ. The feature mapping module ensures that the features of the source and target protocols match correctly during transformation by aligning their hierarchical structures. For example, some features in the source protocol may need to be weighted or transformed to adapt to the format of the target protocol. The aligned features are then normalized to ensure that features from different protocols are on a uniform quantization scale, facilitating subsequent calculations and transformations. Based on the importance of the features and their impact on the protocol transformation, the transformation rule generation module assigns a weight to each normalized feature. This weight represents the degree to which the feature contributes to the protocol transformation process; more critical features are given higher weights and have a greater impact on the transformation result.

[0088] Based on weight allocation and normalized features, dynamic conversion rules are automatically generated. These rules define the conversion priority and mapping relationship between features during the protocol conversion process. For example, a certain feature may have a higher priority in the protocol conversion and need to be processed first, while other features can be processed in later steps. The feature parameters (i.e., conversion rules) are fused with the weight values ​​of the normalized features to form a protocol conversion feature vector containing all the necessary conversion information. This feature vector not only contains the static and dynamic features of the protocol, but also information on how to adapt and convert the protocol according to the conversion rules.

[0089] In an optional embodiment, the protocol conversion feature vector is mapped to a rule space to obtain an initial protocol conversion rule set, including:

[0090] The protocol transformation feature vector is mapped to the initial rule space to obtain the initial rule vector;

[0091] Based on the water information data transmission standard, the effective rule probability of the initial rule vector is obtained;

[0092] Based on the effective rule probability, and based on the adaptive rule pruning method and feature space correction, the effective rule probability is optimized to obtain the initial protocol conversion rule set.

[0093] It should be noted that in the preceding steps, the protocol conversion feature vector has been constructed based on the basic features, related features, and hierarchical relationships of the protocols. This feature vector contains protocol differences, protocol characteristics, and the conversion logic between them. The initial rule space is a predefined space containing rule models for different protocol conversions. Each rule represents a possible operation or process in protocol conversion, such as data format conversion, field mapping, priority adjustment, etc. Mapping the protocol conversion feature vector to the initial rule space means deriving a set of preliminary rules based on the information contained in the feature vector. These rules are initially generated based on the differences and similarities between protocols and are used to describe the conversion relationship from the source protocol to the target protocol. The water information data transmission specification refers to the data transmission requirements and standards in a certain field (such as water conservancy, environmental protection, meteorology, etc.). The water information data transmission specification defines the format, protocol, timing, and other specifications that data should follow when transmitted in the network. These specifications provide a rule framework that must be followed for protocol conversion.

[0094] For each transformation rule in the initial rule vector, their effectiveness may vary. The effective rule probability refers to the probability that a rule can be practically applied under a given water information data transmission specification. Effective rules may be affected by various factors, such as protocol compatibility, network environment, and transmission reliability. For example, in some cases, some rules may violate transmission specifications or fail to work effectively under specific network conditions, while other rules will be more compliant with the specifications and therefore have higher effectiveness. By evaluating and optimizing the effectiveness of each rule, the system can select more effective transformation rules. This process takes into account the application effect of each rule in different scenarios and avoids using rules that do not meet actual transmission requirements or are inefficient. For example, if some rules frequently malfunction or perform poorly during actual transmission, their effectiveness probability will be lowered or even removed from the rule set.

[0095] Adaptive rule pruning is a dynamic adjustment process that automatically adjusts the number and types of rules based on actual needs and usage scenarios. This method allows the system to automatically remove unnecessary or inefficient rules, streamlining the rule set according to different protocol conversion requirements. This pruning method typically relies on changes in data flow, transmission environment, and protocol compatibility, thus flexibly addressing various practical problems. Feature space correction involves adjusting the original feature space to better suit the characteristics of the target protocol. The corrected feature space takes into account the complexity of protocol conversion and the influence of features at different levels. By correcting the feature space, the system can better model protocol features and dynamically adjust features for different protocol types and transmission conditions. After the above optimization process, a streamlined and efficient initial protocol conversion rule set is obtained. This rule set contains conversion rules that conform to actual transmission specifications and requirements, enabling efficient and accurate protocol conversion. The optimized rule set not only removes invalid or inefficient rules but also improves the applicability and effectiveness of the rules, ensuring smooth conversion between protocols.

[0096] In an optional embodiment, the initial protocol conversion rule set is adjusted to obtain the target protocol conversion rule set, including:

[0097] The initial protocol conversion rule set is adjusted by multi-objective collaborative optimization, and a subset of efficient rules is extracted through performance screening to obtain a candidate conversion rule set;

[0098] The candidate transformation rule set is converted into field transformation parameters to obtain the target protocol transformation rule set.

[0099] It's important to note that multi-objective collaborative optimization involves simultaneously considering and optimizing multiple objectives, rather than a single one. Typically, protocol conversion may require optimizing multiple objectives concurrently, such as: efficiency (reducing computational and time overhead during conversion); accuracy (ensuring the converted data is correct in the target protocol); and compatibility (ensuring the converted protocol functions well in the target system). These objectives may conflict or influence each other. Collaborative optimization aims to find a balance among these objectives, ensuring that each objective remains consistent during optimization and collectively improves the overall effectiveness of the protocol conversion. During multi-objective collaborative optimization, the initial protocol... The transformation rule set is adjusted to better support the achievement of multiple objectives. For example, improving certain rules may make the transformation process more efficient and accurate, or sacrifice some details for better overall performance. The initial rule set is evaluated and tested to select rules that are efficient and perform well in practical applications. This process is usually based on certain evaluation metrics, such as: in actual transformation, some rules may not produce any practical effect or may fail to execute successfully, so they need to be removed; some rules may cause unnecessary computational overhead and affect system performance, so inefficient rules need to be removed.

[0100] Some rules may only apply to specific protocols or data formats and are not applicable to the target protocol, so they also need to be filtered out. The result of performance screening is the extraction of a set of efficient and reliable rules that ensure the protocol conversion process is efficient, accurate, and meets the requirements of the target protocol. The efficient rule subset obtained through performance screening is called the candidate conversion rule set. This rule set contains all the rules that meet the requirements and can be further optimized or converted into actual field conversion parameters. Field conversion parameters refer to the specific operations performed on various fields of the protocol (such as fields in data packets, headers, data payloads, etc.) during the protocol conversion process, such as conversion, mapping, formatting, or rearrangement. For example, it may be necessary to map a field A in the source protocol to a field B in the target protocol and adjust its format or unit according to the requirements of the target protocol.

[0101] Each rule in the candidate transformation rule set is specified as an operation instruction, clearly indicating how to transform each protocol field; for example, if the rule is "multiply the value of field X by 2", then the field transformation parameter will explicitly indicate this transformation; the final target protocol transformation rule set is to summarize all field transformation operations to form a complete protocol transformation rule system; it not only includes transformation rules at the protocol level, but also details the transformation at the level of each field, ensuring that the target protocol can correctly receive and parse the data transformed from the source protocol.

[0102] In an optional embodiment, converting the target protocol conversion rule set into device-recognizable transmission instructions and executing them includes:

[0103] The target protocol conversion rule set is parsed based on structured protocol parsing to extract device identifiers, field conversion parameters and transmission timing to obtain the original transmission instructions;

[0104] The water information protocol adapter converts the original transmission commands into a device-compatible protocol format; based on the transmission encryption mechanism, it sends and executes the transmission commands through a secure transmission channel.

[0105] It should be noted that a structured protocol refers to a data exchange protocol organized according to a predefined format, typically including information such as field definitions, field order, and data types. Protocol parsing refers to interpreting and processing the protocol data according to these structured rules. Each device or terminal has a unique identifier, such as a device ID or IP address. In protocol conversion, the device identifier is used to identify the source and destination devices for data exchange to ensure accurate data delivery. Transmission timing refers to the timing requirements that data must follow during transmission. For example, some devices may require data to be sent in a specific order, or some fields may need to be transmitted at specific times. By parsing the structured protocol and extracting the device identifier, field conversion parameters, and transmission timing, the original transmission instructions can be obtained. These instructions are generated based on the target protocol format and specifically specify how to send data, as well as the data order, field content, etc.

[0106] A water information protocol adapter is a protocol conversion component responsible for converting transmission instructions into a protocol format that meets the device's requirements. Device-compatible protocol formats typically include field formats and data structures that the device can understand and accept. The water information protocol adapter's function is to convert the original transmission instructions (target protocol format) into the protocol format required by the device. The adapter usually adjusts the data representation according to the device's needs to ensure that the device can correctly parse and understand the received data. Transmission encryption ensures data security during transmission, preventing data theft or tampering. Encryption mechanisms can encrypt the data content (such as AES encryption) or encrypt the communication channel itself to ensure data confidentiality and integrity.

[0107] A secure transmission channel refers to a secure communication channel established through encryption technology to ensure that the transmission of data from the source device to the target device is confidential and complete. Common secure transmission channels include VPNs, SSL / TLS encrypted channels, and dedicated security protocols. After the data is encrypted and sent through the secure channel, the target device receives the data and performs corresponding operations according to the transmission instructions. These operations may include data storage, control command execution, and response feedback, depending on the protocol conversion requirements.

[0108] In an optional embodiment, dynamic tracking of protocol conversion accuracy, transmission latency, and data integrity is performed to obtain optimized transmission instructions, including:

[0109] The real-time transmission data stream and corresponding conversion logs during the protocol conversion process are obtained. The real-time transmission data stream and conversion logs are cleaned, deduplicated, and formatted to obtain the basic transmission dataset.

[0110] Based on the basic transmission dataset, protocol conversion accuracy features, transmission delay features, and data integrity features are extracted respectively to construct a dynamic tracking model; the dynamic tracking model is used to monitor and record the protocol conversion accuracy features, transmission delay features, and data integrity features in real time to obtain a real-time tracking feature sequence;

[0111] Anomaly identification and correlation analysis are performed on real-time tracking feature sequences to determine feature anomalies and their correlation relationships, resulting in an anomaly correlation map.

[0112] Based on the anomaly correlation graph and the preset transmission quality benchmark, the protocol conversion accuracy constraint, transmission delay constraint and data integrity constraint are determined to obtain a multi-objective constraint parameter set;

[0113] For each abnormal association in the abnormal association graph, the characteristic combination conditions that trigger the relationship are analyzed; based on the characteristic combination conditions and the multi-objective constraint parameter set, the corresponding transmission parameter adjustment scheme is determined, where the transmission parameters include data packet size, retransmission mechanism and encoding method; the transmission parameter adjustment scheme is mapped and stored with the corresponding abnormal association to obtain the strategy library;

[0114] The strategy library is matched and simulated with real-time tracking feature sequences. The optimal adjustment strategy is determined through iterative optimization, and the optimized transmission instructions are obtained.

[0115] It should be noted that real-time data transmission refers to the continuous, real-time flow of data during the protocol conversion process, including all data transmission from the source device to the target device; conversion logs are detailed logs recording the protocol conversion process, which may include data field conversion status, timestamps, error messages, etc.; data cleaning refers to removing duplicate data, abnormal data, and data with inconsistent formats to ensure the accuracy and usability of subsequent analysis; uniform formatting ensures that data follows a consistent structure and type for easy analysis; the cleaned and formatted dataset provides clean and standardized input for subsequent analysis; the accuracy or deviation between the source data and the target protocol format during the protocol conversion process; this may involve the accuracy of field mapping, data type matching, etc.; transmission latency characteristics refer to the time delay from data transmission to reception; this is usually affected by factors such as network conditions and device performance; data integrity characteristics refer to whether the data is intact during transmission, without loss, damage, or modification.

[0116] The dynamic tracking model is a real-time monitoring and recording model for data changes. It tracks protocol conversion accuracy, latency, and integrity characteristics, recording this data in real time for subsequent analysis. With the help of the dynamic tracking model, protocol conversion accuracy, latency, and data integrity are continuously monitored and recorded chronologically, generating a series of real-time feature sequences. Analysis of these feature sequences identifies unexpected values, which may be due to network fluctuations, equipment failures, or other factors. The model also analyzes the correlations between different features to identify which feature changes might be caused by a common reason; for example, data integrity issues may be related to increased transmission latency. Based on this analysis, an anomaly correlation graph is generated between the features. This graph shows which features are interrelated and which features often occur together under abnormal conditions, helping to understand the root cause of the anomalies.

[0117] The preset standards or requirements may include ideal or tolerable ranges for protocol conversion accuracy, latency, and integrity; protocol conversion accuracy constraints, transmission latency constraints, and data integrity constraints define the fault tolerance standards or expected values ​​during the protocol conversion and data transmission processes; for example, the error in protocol conversion may need to be controlled within a certain range, the transmission latency should not exceed a certain threshold, and data integrity must be guaranteed to be 100%; the multi-objective constraint parameter set is a set of various constraints comprehensively formulated based on anomaly correlation maps and transmission quality benchmarks to ensure that the transmission quality meets the predetermined requirements; it analyzes which characteristic combinations will lead to the occurrence of anomaly correlations; for example, excessive transmission latency may lead to data integrity problems, or low protocol conversion accuracy may lead to increased latency;

[0118] Based on triggering conditions and constraint sets, corresponding transmission parameter adjustment schemes are formulated to optimize data packet size, retransmission mechanisms, and encoding methods. Adjusting the data packet size may help improve transmission efficiency or reduce packet loss rate. It is determined whether the retransmission mechanism needs to be enhanced, such as increasing the number of retransmissions or adopting different retransmission strategies. Appropriate encoding methods, such as compression encoding or error correction encoding, are selected to help improve the reliability and efficiency of data transmission. The strategy library is a repository of transmission parameter adjustment schemes, with each record associated with a certain anomaly. When a corresponding anomaly occurs, the transmission parameters can be optimized according to the adjustment schemes provided by the strategy library. Based on the real-time monitored feature sequences, appropriate adjustment strategies are matched and simulation tests are conducted to evaluate the effectiveness of the strategies. Through continuous optimization of the strategies (e.g., multiple simulations and adjustments), the optimal transmission adjustment strategy is found. The final optimized transmission command is based on the output of the strategy library and the dynamic tracking model, aiming to improve the efficiency, accuracy, and reliability of data transmission.

[0119] In an optional embodiment, based on an anomaly correlation graph and a preset transmission quality benchmark, protocol conversion accuracy constraints, transmission delay constraints, and data integrity constraints are determined to obtain a multi-objective constraint parameter set, including:

[0120] The influence weights of each characteristic anomaly type are extracted from the anomaly correlation graph, where the influence weights represent the degree of interference of the anomaly type on the overall transmission quality.

[0121] The preset transmission quality benchmark is dynamically adjusted based on the influence weight to determine the maximum allowable deviation in protocol conversion accuracy, the longest transmission delay threshold, and the minimum standard for data integrity.

[0122] By integrating the maximum deviation in protocol conversion accuracy, the longest threshold for transmission delay, and the minimum standard for data integrity, a multi-objective constraint parameter set is obtained.

[0123] It should be noted that in the transmission quality graph, abnormal correlations between certain features may have a greater impact on the overall transmission quality. For example, excessive transmission latency may directly lead to data loss or decreased transmission accuracy, thus its impact weight may be relatively large. The magnitude of the impact weight measures the degree of interference of the abnormal feature on transmission quality. For instance, if a feature (such as protocol conversion accuracy) is abnormal, it may affect the overall transmission quality, but if its abnormality is weakly correlated with other features (such as data integrity), then its impact weight is smaller. The calculation of impact weights can be based on statistical models, machine learning methods, or expert ratings, ultimately quantifying the actual impact of each anomaly type on transmission quality. The transmission quality benchmark refers to the ideal standard that transmission quality should achieve in the absence of anomalies. Typically, these benchmark values ​​are preset, representing the optimal or allowable range of features such as protocol conversion accuracy, transmission latency, and data integrity. When the system detects anomalies in certain features, these benchmark values ​​must be adjusted according to the impact weight of the anomaly type. In other words, the benchmark values ​​are no longer static but dynamically adjusted according to the degree of interference from the anomaly.

[0124] The maximum deviation in protocol conversion accuracy refers to the maximum allowable error range during protocol conversion. For example, if a slight deviation in protocol conversion accuracy has little impact on overall transmission quality, the tolerance for the maximum deviation can be appropriately relaxed. The maximum transmission delay threshold defines the maximum data transmission time. If the transmission delay is too long, it may lead to packet loss or timeout, so a reasonable maximum delay threshold needs to be set. Based on the analysis of influence weights, if the interference of delay anomalies on transmission quality is small, the delay tolerance range can be appropriately extended. The minimum standard for data integrity refers to the level of integrity that data must maintain during transmission. If certain characteristics (such as delay or protocol conversion accuracy) are abnormal, it may be necessary to reduce the requirements for data integrity and allow a certain tolerance range. The adjustment standards for these three characteristics—protocol conversion accuracy, transmission delay, and data integrity—may affect each other. For example, higher protocol conversion accuracy may increase delay, while relaxing the delay standard may affect data integrity. Therefore, a balance must be found among multiple objectives to ensure that each characteristic is within a reasonable constraint range. Based on the multi-objective constraint parameter set, the optimal transmission parameter adjustment scheme is selected to ensure that the overall transmission quality is within the reasonable range of each characteristic.

[0125] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A protocol conversion method of an Internet of Things water information heterogeneous data transmission protocol gateway, characterized by, The method comprises the following steps: Obtain the heterogeneous protocol characteristics and historical transmission data of the Internet of Things water information equipment, construct a protocol characteristic time sequence graph, and generate a protocol difference matrix based on the protocol characteristic time sequence graph; The generation of the protocol difference matrix based on the protocol characteristic time sequence graph comprises the following steps: Analyze the node attributes and edge association information of the protocol characteristic time sequence graph, extract a protocol static feature vector and a time sequence dynamic association strength, determine a feature comparison dimension and an association comparison dimension, obtain a feature reference set and an association reference set, and determine a feature reference set and an association reference set; Establish a deviation quantization rule for the feature comparison dimension and a variation quantization rule for the association comparison dimension, compare the protocol static feature vector item by item based on the feature reference set, obtain a feature deviation degree, compare the time sequence dynamic association strength based on the association reference set, obtain an association variation degree, group the comparison dimensions, and obtain a same-dimension difference data set; Take the feature comparison dimension as the row dimension of the matrix and the association comparison dimension as the column dimension of the matrix to construct a matrix framework, map the feature deviation degree and the association variation degree in the same-dimension difference data set to the corresponding positions of the matrix framework, determine a difference level division standard, label the difference states of the elements at each position, and obtain an initial protocol difference matrix; Collect real-time protocol interaction data, calculate an adjustment coefficient based on the element distribution characteristics of the initial protocol difference matrix, update the numerical values of the feature deviation degree and the association variation degree based on the adjustment coefficient, synchronously update the difference state labels of the elements at each position, and obtain a protocol difference matrix; Extract a cross-protocol conversion feature based on the protocol difference matrix, construct a heterogeneous protocol conversion model, and obtain a protocol conversion feature vector; The extraction of the cross-protocol conversion feature based on the protocol difference matrix, the construction of the heterogeneous protocol conversion model, and the obtaining of the protocol conversion feature vector comprise the following steps: Statistically analyze the occurrence frequencies of each difference attribute in different protocol pairs in the protocol difference matrix, filter high-frequency difference attributes as basic features, analyze the co-occurrence relationships among the basic features, obtain association features, hierarchically divide the basic features and the association features, mark the levels to which each feature belongs and the dependency relationships among the levels, integrate the basic features, the association features, and the level dependency relationships, and obtain a cross-protocol conversion feature set; Based on the cross-protocol conversion feature set, construct a heterogeneous protocol conversion model, take the cross-protocol conversion feature set as the feature source of the input layer of the model, set a feature mapping module and a conversion rule generation module, wherein the feature mapping module is used for feature dimension adaptation, and the conversion rule generation module is used for generating conversion logic based on the adapted features. The feature mapping module receives the cross-protocol conversion feature set, performs dimensional alignment based on the hierarchical attributes of each feature, maps different hierarchical features to a unified dimensional space, and obtains normalized features; the conversion rule generation module performs weight allocation on the normalized features; and a dynamic conversion rule is generated based on the weight allocation result and the normalized features, wherein the dynamic conversion rule includes conversion priorities and mapping relationships between features; the feature parameters of the dynamic conversion rule are vector fused with the weight values of the normalized features, and a protocol conversion feature vector is obtained; The protocol conversion feature vector is mapped to a rule space to obtain an initial protocol conversion rule set; The initial protocol conversion rule set is adjusted to obtain a target protocol conversion rule set; The target protocol conversion rule set is converted into a transmission instruction recognizable by a device and executed; The protocol conversion accuracy, transmission delay, and data integrity are dynamically tracked to obtain an optimized transmission instruction; The protocol conversion accuracy, transmission delay, and data integrity are dynamically tracked to obtain an optimized transmission instruction, including: Real-time transmission data streams and corresponding conversion logs in the protocol conversion process are obtained, and the real-time transmission data streams and conversion logs are processed through cleaning, deduplication, and format unification to obtain a basic transmission data set; Based on the basic transmission data set, protocol conversion accuracy features, transmission delay features, and data integrity features are extracted respectively to construct a dynamic tracking model; the protocol conversion accuracy features, transmission delay features, and data integrity features are monitored and recorded in sequence in real time through the dynamic tracking model to obtain a real-time tracking feature sequence; The real-time tracking feature sequence is subjected to outlier identification and correlation analysis to determine feature outliers and abnormal correlation relationships, and an abnormal correlation graph is obtained; Based on the abnormal correlation graph and based on a preset transmission quality benchmark, protocol conversion accuracy constraints, transmission delay constraints, and data integrity constraints are determined to obtain a multi-objective constraint parameter set; For each abnormal correlation relationship in the abnormal correlation graph, the feature combination conditions triggering the relationship are analyzed; based on the feature combination conditions and the multi-objective constraint parameter set, a corresponding transmission parameter adjustment scheme is determined, wherein the transmission parameters include data packet size, retransmission mechanism, and encoding mode; the transmission parameter adjustment scheme and the corresponding abnormal correlation relationship are mapped and stored to obtain a strategy library; The strategy library and the real-time tracking feature sequence are matched and simulated to determine an optimal adjustment strategy through iterative optimization to obtain an optimized transmission instruction. 2.The protocol conversion method of the Internet of Things water information heterogeneous data transmission protocol gateway according to claim 1, characterized in that, A protocol feature time sequence graph is constructed, including: The heterogeneous protocol features and historical transmission data are matched and associated to obtain effective correlation pairs; based on the correspondence between protocol attribute categories and data attribute categories, the same attribute correlation data set is obtained; Based on the effective association pairs in the same attribute association dataset, the core attributes and time sequence attributes of the graph nodes are set; based on the association strength of the heterogeneous protocol features and historical transmission data, the association weight and time sequence constraint condition of the graph edge are defined; based on the node attributes, edge association weight and time sequence constraint condition, the time sequence graph basic framework is constructed; the transmission data time sequence association path between different nodes is established, the effective association pairs are bound in time sequence order and nodes and edges to obtain an initial protocol feature time sequence graph; The real-time transmission data of the Internet of Things water information equipment is collected, the graph adaptation coefficient is calculated based on the node association condition of the initial protocol feature time sequence graph and the data transmission efficiency, the graph adaptation coefficient is compared with the preset adaptation threshold to obtain an adaptation comparison result; based on the adaptation comparison result, the current adaptation state of the initial protocol feature time sequence graph is determined, and if the adaptation state does not meet the preset requirement, the core attributes and time sequence attributes of the initial protocol feature time sequence graph nodes are updated; the association weight and time sequence constraint condition of the initial protocol feature time sequence graph edge are adjusted to obtain a protocol feature time sequence graph. 3.The protocol conversion method of the Internet of Things water information heterogeneous data transmission protocol gateway according to claim 2, characterized in that, Map the protocol conversion feature vector to the rule space to obtain an initial protocol conversion rule set, including: Map the protocol conversion feature vector to the initial rule space to obtain an initial rule vector; Based on the water information data transmission specification, the effective rule probability of the initial rule vector is obtained; Based on the effective rule probability, and based on the adaptive rule pruning method and feature space correction, the effective rule probability is optimized to obtain the initial protocol conversion rule set. 4.The protocol conversion method of the Internet of Things water information heterogeneous data transmission protocol gateway according to claim 3, characterized in that, Adjust the initial protocol conversion rule set to obtain a target protocol conversion rule set, including: Adjust the initial protocol conversion rule set using multi-objective collaborative optimization, extract a high-efficiency rule subset through efficiency screening to obtain a candidate conversion rule set; Convert the candidate conversion rule set into field conversion parameters to obtain the target protocol conversion rule set. 5.The protocol conversion method of the Internet of Things water information heterogeneous data transmission protocol gateway according to claim 4, characterized in that, Convert the target protocol conversion rule set into a transmission instruction recognizable by the equipment and execute it, including: Based on structured protocol analysis, the target protocol conversion rule set is analyzed to extract the device identifier, field conversion parameter and transmission time sequence to obtain the original transmission instruction; Based on the water information protocol adapter, the original transmission instruction is converted into a protocol format compatible with the equipment; based on the transmission encryption mechanism, the transmission instruction is sent through a secure transmission channel and executed. 6.The protocol conversion method of the Internet of Things water information heterogeneous data transmission protocol gateway according to claim 5, characterized in that, Based on the abnormal association graph and based on the preset transmission quality benchmark, the protocol conversion precision constraint, the transmission time delay constraint and the data integrity constraint are determined to obtain a multi-objective constraint parameter set, including: Extract the influence weight of each feature abnormal type from the abnormal association graph, wherein the influence weight represents the interference degree of the abnormal type on the overall transmission quality; Based on the influence weight, the preset transmission quality benchmark is dynamically adjusted to determine the allowed protocol conversion precision maximum deviation, the transmission time delay longest threshold and the data integrity minimum standard; Integrate the protocol conversion precision maximum deviation, the transmission time delay longest threshold and the data integrity minimum standard to obtain the multi-objective constraint parameter set.

Images