Multi-source data-based multi-agent total nitrogen emission load allocation method and system

Abstract

The application discloses a multi-source data-based multi-agent total nitrogen emission load distribution method and system, relates to the field of total nitrogen load distribution of a river basin, and comprises the following steps: collecting and processing multi-source heterogeneous data, and outputting a multi-source standardized data set; using a Bayesian update engine to infer the hidden credit status and governance ability level of each pollution discharge agent, and outputting a dynamic agent credit and ability portrait set; based on the dynamic agent credit and ability portrait set, distributing the flexible capacity part of the total allowable total nitrogen emission of the river basin to each pollution discharge agent through a market mechanism, combining the basic capacity to form a differentiated load distribution scheme draft, and associating and mapping the distribution result in the differentiated load distribution scheme draft with the evaluation information in the dynamic agent credit and ability portrait set. The application realizes the full-process automation, precision and traceable load distribution management from data collection, dynamic evaluation, intelligent distribution to credible execution and closed-loop feedback.

Description

Technical Field

[0001] This invention relates to the field of total nitrogen load allocation in watersheds, and in particular to a multi-entity total nitrogen emission load allocation method and system based on multi-source data. Background Technology

[0002] In the field of total nitrogen load allocation in watersheds, achieving efficient allocation is a key challenge. Traditional methods mainly rely on optimization algorithms based on water environment capacity models and mathematical programming theory. For example, by constructing cost-benefit or equity-efficiency objective functions and combining them with hydrological and water quality constraints, a theoretically optimal emission limit is calculated for each polluter. This drives a central calculation model, providing an important theoretical analysis tool for policy making. A static allocation scheme is generated through a one-time calculation, and then implementation is promoted through administrative means.

[0003] The main challenge of traditional methods is their inability to dynamically respond to the differentiated inherent attributes of polluting entities, and the disconnect between allocation decisions and enforcement supervision. Existing technologies typically treat polluting entities as homogeneous and static entities, making static allocations based on limited preset parameters. This fails to effectively perceive and respond to their actual behavioral patterns and credit changes, making it difficult for the allocation mechanism itself to continuously incentivize or constrain the behavior of entities. As a static administrative order, the allocation scheme relies on independent manual supervision processes for execution, resulting in a separation between calculation and execution. This separation hinders dynamic adjustments based on actual performance and affects the transparency and credibility of the execution process. There is a lack of an integrated technical system that can combine differentiated allocation, reliable automatic execution, and closed-loop feedback. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a multi-subject total nitrogen emission load allocation method based on multi-source data to solve the technical problems of disconnect between decision-making and execution, rigid mechanisms, and lack of dynamic feedback loops in the existing watershed total nitrogen load allocation.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a multi-subject total nitrogen emission load allocation method based on multi-source data, which includes collecting and processing multi-source heterogeneous data and outputting a multi-source standardized dataset.

[0008] Using a Bayesian update engine, we infer the hidden credit status and governance capacity level of each polluting entity, and output a dynamic entity credit and capacity profile set.

[0009] Based on a dynamic entity credit and capability profile set, the flexible capacity portion of the total allowable nitrogen emission in the basin is allocated to each polluting entity through market mechanisms, and a draft differentiated load allocation scheme is formed in combination with the basic capacity.

[0010] The allocation results in the draft differentiated load allocation scheme are correlated and mapped with the assessment information in the dynamic subject credit and capability profile set to generate differentiated smart contract execution parameters. The differentiated smart contract execution parameters are then deployed to the blockchain network to form a set of smart contracts with differentiated parameters.

[0011] By acquiring new multi-source standardized datasets through off-chain trusted data sources, a set of smart contracts with differentiated parameters on the blockchain network is triggered to automatically execute quota management and generate on-chain trusted execution records.

[0012] By combining on-chain trusted execution records with new multi-source standardized datasets, we update the dynamic entity credit and capability profile set and adjust the set of smart contracts with differentiated parameters.

[0013] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method involves: collecting and processing multi-source heterogeneous data to output a multi-source standardized dataset, including the following steps:

[0014] Based on the spatiotemporal alignment of the collected multi-source heterogeneous data, a spatiotemporal aligned data set is generated. The spatiotemporal aligned data set is then cleaned to obtain a multi-source standardized dataset.

[0015] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method includes the following steps: using a Bayesian update engine to infer the hidden credit status and governance capacity level of each polluting entity, and outputting a dynamic entity credit and capacity profile set.

[0016] Based on multi-source standardized datasets, a hidden state hypothesis space with multiple potential behavioral patterns is defined for each sewage discharge entity. Prior probability weights are configured for the hidden state hypothesis space according to industry benchmarks and historical records to form an initial state probability distribution and construct a Bayesian update engine.

[0017] The multi-source standardized dataset is used as the input of the dynamic evidence stream into the Bayesian update engine. Within the cognitive framework of the pollution subject's behavior, the correlation strength between the dynamic evidence stream and each hidden state is analyzed to generate a multi-dimensional likelihood correlation strength spectrum.

[0018] The cognitive weight allocation of the initial state probability distribution is adjusted based on the multidimensional likelihood correlation strength spectrum, and the belief state probability distribution reflecting the degree of support of the latest evidence is evolved through iterative updates.

[0019] Based on the aforementioned belief state probability distribution, the credit representation mapping rule is used to condense high-dimensional probabilistic beliefs into low-dimensional operable credit representation vectors.

[0020] Based on the probability distribution of belief states, the robustness measure of the operation mode that oscillates in tandem with the high integrity state belief in the hidden state hypothesis space is analyzed from the governance facility operation data fragments of the multi-source standardized dataset using the capability validity extraction rule.

[0021] The robustness measure of the operating mode is compared with the preset efficiency threshold spectrum to determine the treatment capacity level of each pollutant.

[0022] The probability distribution of each pollutant's belief state, credit representation vector, and governance capacity level are topologically linked and integrated to form a dynamic profile of the pollutant's credit and capacity.

[0023] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method includes the following steps: Based on a dynamic entity credit and capacity profile set, the elastic capacity portion of the total allowable total nitrogen emission in the watershed is allocated to each polluting entity through a market mechanism.

[0024] Based on the watershed water environment model and water quality targets, the total allowable discharge of total nitrogen in the watershed is calculated and divided into basic discharge capacity and flexible regulation capacity.

[0025] Based on a dynamic entity credit and capability profile and evaluation criteria, a differentiated capacity application and evaluation mechanism is constructed for multiple rounds of sealed submission of allocation applications;

[0026] In the differentiated capacity application and evaluation mechanism, each polluting entity submits a sealed allocation application based on its demand for flexible adjustment capacity and the application priority determined by its own dynamic entity credit and capability profile.

[0027] Based on the sealed allocation application results in the differentiated capacity application and evaluation mechanism, the flexible adjustment capacity is allocated to the top-ranked sewage discharge entities according to the application ranking adjusted by the evaluation priority coefficient.

[0028] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method includes the following steps: Forming a draft differentiated load allocation scheme by combining the baseline capacity.

[0029] The results of the flexible adjustment capacity allocation are summed and aggregated with the pre-set basic emission capacity to generate the final quota of basic emission capacity and the allocated flexible adjustment capacity for each pollutant, forming a draft of the differentiated load allocation scheme.

[0030] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method involves: associating and mapping the allocation results in the draft differentiated load allocation scheme with the evaluation information in the dynamic entity credit and capability profile set to generate differentiated smart contract execution parameters, including the following steps:

[0031] Based on the draft of the differentiated load allocation scheme, the final quota value allocated to each pollutant in the draft of the differentiated load allocation scheme is extracted to form a final quota allocation list.

[0032] Based on the dynamic subject credit and capability profile set, the credit representation vector and governance capability level corresponding to each polluting subject in the dynamic subject credit and capability profile set are extracted to form a credit and capability attribute set of credit representation vector and governance capability level.

[0033] Based on the preset parameter mapping rules, the credit representation vector and the credit and capability attribute set of governance capability level are mapped to smart contract function parameters.

[0034] Based on the final quota allocation list and smart contract function parameters, a smart contract parameter configuration file is generated for each polluting entity. The smart contract parameter configuration files of the polluting entities are aggregated to form a differentiated set of smart contract execution parameters.

[0035] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method involves: deploying differentiated smart contract execution parameters to a blockchain network to form a set of smart contracts with differentiated parameters, including the following steps:

[0036] Based on a differentiated set of smart contract execution parameters, smart contract logic is customized for each polluting entity to generate an individual smart contract deployment package;

[0037] Individual smart contract deployment packages are submitted to the permissioned consortium blockchain network. After verification by network consensus, an independent smart contract account instance is created and activated on the blockchain for each polluting entity.

[0038] The smart contract account instances created and activated on the blockchain are aggregated to form a set of smart contracts with differentiated parameters.

[0039] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method includes the following steps: obtaining a new multi-source standardized dataset through a trusted off-chain data source, triggering a set of smart contracts with differentiated parameters on the blockchain network to automatically execute quota management, and generating a trusted on-chain execution record.

[0040] The off-chain trusted data source collects and processes data from the online monitoring system of polluting enterprises and other sources according to a preset cycle, generating the latest multi-source standardized dataset.

[0041] The latest multi-source standardized dataset is signed and transmitted to the blockchain network in the form of oracle transactions. The set of smart contracts with differentiated parameters on the blockchain network receives the oracle transactions, automatically parses the emission data in the latest multi-source standardized dataset, and executes quota verification, transaction and reward / penalty operations according to the built-in logic of the contract, generating a trusted on-chain execution record.

[0042] As a preferred embodiment of the multi-entity total nitrogen emission load allocation method based on multi-source data described in this invention, the method includes the following steps: updating the dynamic entity credit and capability profile set by combining on-chain trusted execution records with a new multi-source standardized dataset, and adjusting the set of smart contracts with differentiated parameters:

[0043] The on-chain trusted execution records and the new multi-source standardized dataset are used as a new evidence stream and input into the Bayesian update engine. Based on the new evidence stream, the original belief state probability distribution is iteratively updated to generate an updated dynamic subject credit and capability profile set.

[0044] Based on the updated dynamic entity credit and capability profile set, the execution logic of the set of smart contracts with differentiated parameters on the blockchain network is dynamically reconfigured through parameter adjustment transactions, thereby adjusting the set of smart contracts with differentiated parameters.

[0045] Secondly, the present invention provides a multi-entity total nitrogen emission load allocation system based on multi-source data, including a processing module that collects and processes multi-source heterogeneous data and outputs a multi-source standardized dataset.

[0046] The inference module uses a Bayesian update engine to infer the hidden credit status and governance capacity level of each polluting entity, and outputs a dynamic entity credit and capacity profile set.

[0047] The allocation module, based on a dynamic entity credit and capability profile set, allocates the elastic capacity portion of the total allowable nitrogen emissions in the basin to each polluting entity through market mechanisms, and forms a draft differentiated load allocation scheme in combination with the basic capacity;

[0048] The mapping module associates and maps the allocation results in the draft of the differentiated load allocation scheme with the evaluation information in the dynamic subject credit and capability profile set, generates differentiated smart contract execution parameters, and deploys the differentiated smart contract execution parameters to the blockchain network to form a set of smart contracts with differentiated parameters.

[0049] The execution module obtains new multi-source standardized datasets through off-chain trusted data sources, triggers a set of smart contracts with differentiated parameters on the blockchain network to automatically execute quota management, and generates on-chain trusted execution records;

[0050] The update module updates the dynamic entity credit and capability profile set by combining on-chain trusted execution records with a new multi-source standardized dataset and adjusting the set of smart contracts with differentiated parameters.

[0051] The beneficial effects of this invention are as follows: By aggregating and processing multi-source heterogeneous data to form a standardized dataset, and using a Bayesian update engine to dynamically infer the credit status and governance capacity level of the polluting entity, a dynamic entity credit and capacity profile set is generated. Combined with the basic capacity and the elastic capacity allocated through the market mechanism, a draft of a differentiated load allocation scheme is formed. The allocation results and evaluation information are mapped into differentiated smart contract execution parameters, which are deployed to the blockchain network to form a set of smart contracts with differentiated parameters. The smart contracts are automatically executed for quota management through off-chain trusted data sources, generating on-chain trusted execution records. These records are used to update the profile set and adjust the smart contracts with the new dataset, thereby realizing fully automated, precise, and traceable load allocation management from data collection, dynamic evaluation, intelligent allocation to trusted execution and closed-loop feedback. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a flowchart of a multi-entity total nitrogen emission load allocation method based on multi-source data.

[0054] Figure 2 This is a schematic diagram of a multi-entity total nitrogen emission load allocation system based on multi-source data. Detailed Implementation

[0055] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0056] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0057] Secondly, the term "one embodiment" or "example" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the invention. The appearance of an embodiment in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0058] Reference Figures 1-2 This is one embodiment of the present invention, which provides a multi-entity total nitrogen emission load allocation method based on multi-source data, including the following steps:

[0059] S1. Collect and process multi-source heterogeneous data, and output multi-source standardized datasets.

[0060] S1.1 Based on the spatiotemporal alignment processing of the collected multi-source heterogeneous data, a spatiotemporal aligned data set is generated. The spatiotemporal aligned data set is then cleaned to obtain a multi-source standardized dataset.

[0061] Furthermore, multi-source heterogeneous data is collected from various sources, including online monitoring systems of polluting enterprises, supervisory monitoring platforms of ecological and environmental departments, satellite remote sensing data centers, power grid companies, drone patrol systems, and pollutant discharge permit management platforms. The collected multi-source heterogeneous data undergoes spatiotemporal alignment processing, assigning a standardized timestamp and geographic coordinates to each data record. For example, the time base of all data records is calibrated to Beijing time, and the geographic coordinates are unified to the WGS84 coordinate system, generating a spatiotemporally aligned dataset. This dataset is then cleaned to identify and address missing values, outliers, and format errors. For instance, missing total nitrogen monitoring concentrations are interpolated using the average of adjacent time points at the same monitoring point. Flow rates significantly exceeding reasonable physical ranges are considered outliers and removed, generating a cleaned dataset. Finally, various physical quantities in the cleaned dataset are standardized, converting data from different sources and dimensions into a unified numerical range. For example, a maximum-minimum normalization method is used to map all total nitrogen concentration values ​​to between zero and one, completing the construction of a multi-source standardized dataset.

[0062] S2. Using a Bayesian update engine, infer the hidden credit status and governance capacity level of each polluting entity, and output a dynamic entity credit and capacity profile set.

[0063] S2.1 Based on multi-source standardized datasets, define a hidden state hypothesis space for each sewage discharge entity with multiple potential behavioral patterns. Configure prior probability weights for the hidden state hypothesis space according to industry benchmarks and historical records to form an initial state probability distribution and build a Bayesian update engine.

[0064] Furthermore, a hidden state hypothesis space is defined for each polluting entity, consisting of three types: honest compliance, insufficient technical capability, and strategic violation. This definition is based on the induction of behavioral patterns from multi-source standardized data and domain knowledge. Based on the statistical information of various entity proportions in industry benchmark reports, combined with the polluting entity's historical compliance records and emission performance over a period of time, a priori probability weight is assigned to each state in the hidden state hypothesis space. For example, for entities with generally good performance in industry benchmarks and no historical violations, a higher priori probability weight can be assigned to the honest compliance state. The hidden state hypothesis space with configured prior probability weights is then formally constructed as the initial state probability distribution for each polluting entity.

[0065] Specifically, the core computational framework of the Bayesian update engine is built based on the probability distribution of the initial state. The Bayesian update engine can receive dynamic evidence input and perform probability updates according to Bayes' theorem, establishing a quantitative and updatable probabilistic cognitive starting point for each polluting entity. It transforms subjective and vague initial judgments of credit and ability into prior probability distributions, so that the entire assessment process starts from an evidence-based initial belief, rather than from zero or arbitrary assumptions, thereby enhancing the scientific nature and interpretability of the assessment process.

[0066] S2.2. Using the multi-source standardized dataset as the input to the Bayesian update engine for dynamic evidence stream, within the cognitive framework of the polluting subject's behavior, the correlation strength between the dynamic evidence stream and each hidden state is analyzed, and a multi-dimensional likelihood correlation strength spectrum is generated.

[0067] Furthermore, a multi-source standardized dataset, including online monitoring concentrations, treatment facility operating parameters, and enterprise production data, is continuously input into the Bayesian update engine as a dynamic evidence stream organized in a time series. Within a pre-defined cognitive framework of the polluter's behavior, this framework clarifies the typical evidence patterns corresponding to each hidden state. For example, the evidence pattern corresponding to the integrity and compliance state might include emission concentrations consistently below standard limits with minimal fluctuations, key parameters of treatment facilities operating within their rated ranges, and a high degree of matching between production load and the energy consumption of treatment facilities. The Bayesian update engine analyzes each batch of data combinations in the dynamic evidence stream to obtain the conditional probability, or likelihood probability, of that data combination appearing in each hidden state within the hidden state hypothesis space.

[0068] Specifically, by integrating the likelihood probabilities of all evidence dimensions, a multi-dimensional likelihood correlation strength spectrum reflecting the correlation strength between dynamic evidence flow and each hidden state is generated. This transforms multi-dimensional real-time observation data into standardized quantitative support for the probability of hidden states. By defining state-evidence correlation patterns and calculating likelihood, the raw data flow, which is difficult to interpret directly, is transformed into evidence language that can be directly used to update probabilistic beliefs, thus realizing the key data preprocessing and transformation steps for dynamic cognition.

[0069] S2.3. Adjust the cognitive weight allocation of the initial state probability distribution based on the multi-dimensional likelihood correlation strength spectrum, and iteratively update and evolve the belief state probability distribution that reflects the degree of support of the latest evidence.

[0070] Furthermore, the Bayesian update engine evaluates the received multi-dimensional likelihood correlation strength spectrum in conjunction with the current belief state probability distribution (initially the initial state probability distribution), strictly following Bayes' theorem. It uses the multi-dimensional likelihood correlation strength spectrum as the likelihood function and the current belief state probability distribution as the prior probability, and obtains a new posterior probability distribution through Bayes' formula. The posterior probability distribution is the updated belief state probability distribution. Each new dynamic evidence stream input triggers such an iterative update calculation, causing the belief state probability distribution to continuously evolve over time and with the accumulation of evidence.

[0071] Specifically, it enables continuous, dynamic, and quantitative inference of the credit and capacity status of polluting entities, breaking away from the rigid cycle of traditional periodic assessments. It achieves real-time or near-real-time perception of the entity's status. The mathematical properties of Bayesian updates ensure that the assessment results can smoothly integrate new and old evidence. It can respond to sudden abnormal events without fluctuating drastically due to single data fluctuations, forming a stable and sensitive probability evolution trajectory. This provides robust cognition for decision-making that reflects the latest situation while taking into account historical performance.

[0072] S2.4 Based on the probability distribution of belief states, the credit representation mapping rule is used to condense high-dimensional probabilistic beliefs into low-dimensional operable credit representation vectors.

[0073] Furthermore, based on the updated belief state probability distribution, a pre-defined credit representation mapping rule is applied. This rule defines how to map a high-dimensional belief state probability distribution into a low-dimensional, more business-interpretive credit representation vector. A typical credit representation mapping rule extracts the posterior probability of the honest and compliant state from the belief state probability distribution, and may combine it with the probabilities of other states for weighting or transformation to generate a scalar value as a comprehensive credit score, constituting the core element of the credit representation vector. Other dimensions, such as the probability of strategic violations, are further extracted to represent risk propensity, collectively forming a multi-dimensional credit representation vector.

[0074] Specifically, the complex internal probability inference results are transformed into standardized and concise credit indicators that can be directly understood and used by external decision-making processes (such as market mechanisms and contract terms). Dimensionality reduction and interface-based processing bridge the semantic gap between probability model outputs and business rule inputs, enabling dynamic evaluation based on advanced statistical inference to be seamlessly integrated into specific operation and management processes.

[0075] S2.5 Based on the probability distribution of belief states, the robustness measure of the operation mode that oscillates in tandem with the high integrity state belief in the hidden state hypothesis space is analyzed from the governance facility operation data fragments of the multi-source standardized dataset using the capability validity extraction rule.

[0076] Furthermore, based on the probability distribution of belief states, particularly the posterior probability sequence of the integrity and compliance state, data segments directly related to the operation of the treatment facilities are extracted from multi-source standardized datasets. These include time-series data on wastewater treatment unit influent flow, chemical dosage, energy consumption, and key water quality parameters (such as pH and dissolved oxygen). Capability validity extraction rules are used to analyze these operational data segments. The core of these rules is to identify operational patterns in the operational data that are synchronous or lagging in time with stable statistical characteristics of high integrity belief states (i.e., periods with high or rising posterior probabilities of integrity and compliance). For example, during periods of high integrity belief states, the treatment facilities may exhibit lower unit energy consumption volatility, more stable control accuracy of key parameters, and faster anomaly recovery response.

[0077] Specifically, the capability validity extraction rule quantifies and integrates these characteristics into a comprehensive scalar indicator, namely the robustness measure of the operational mode, through time series alignment, correlation analysis, and stability index calculation methods. It does not assess technical capabilities in isolation, but links technical operational performance with changes in credit probability. It believes that truly effective governance capabilities will exhibit robust and efficient characteristic patterns during periods of high credit, avoiding the possibility of being misled by strategic compliance (such as only meeting standards during sampling) when judging capabilities solely from technical parameters. It more accurately extracts data characteristics that reflect true and continuous governance effectiveness.

[0078] S2.6. Compare the robustness measurement of the operating mode with the preset efficiency threshold spectrum to determine the treatment capacity level of each sewage discharge entity.

[0079] Furthermore, the robustness measure of the operating mode obtained from the capability validity extraction rule is compared with a preset efficiency threshold spectrum. The efficiency threshold spectrum is a series of threshold intervals predefined based on industry technical specifications, best practice data of similar facilities, or statistical analysis (such as quantile division) of the historical operating mode robustness measures of all polluting entities. Each threshold interval corresponds to a governance capability level, such as high efficiency, stable, basic compliance, and need for improvement. The governance capability level of each polluting entity is determined by which preset efficiency threshold spectrum interval the operating mode robustness measure value of each polluting entity falls into.

[0080] Specifically, continuous technical performance metrics are transformed into discrete, easy-to-understand, and easy-to-operate classification labels. Through preset and adjustable performance threshold spectra, managers can flexibly define specific technical thresholds for levels such as high efficiency and stability based on management objectives, technological development levels, or policy requirements. This ensures that the capability assessment results are based on objective data and can also adapt to the needs of management classification and policy implementation, thus achieving an effective bridge between technical metrics and management classification.

[0081] S2.7. Integrate the probability distribution of belief state, credit representation vector and governance capacity level of each polluting entity through topological association to form a dynamic entity credit and capacity profile set.

[0082] Furthermore, for each polluting entity, the belief state probability distribution output by the Bayesian update engine, the credit representation vector generated by the credit representation mapping rules, and the governance capacity level calibrated by the efficiency threshold spectrum comparison are structurally integrated. This integration is not simply storing the three independent results side by side, but rather establishing the intrinsic relationship between them. For example, in the dynamically shaped entity credit and capacity profile set, not only are the values ​​of the credit representation vector and the labels of the governance capacity level recorded, but the entire picture of the belief state probability distribution that produced these results is also preserved. It may also indicate which state(s) in the belief state probability distribution the credit representation vector mainly comes from, as well as the correlation strength between the robustness measure of the operating mode on which the governance capacity level is based and the belief of high integrity.

[0083] Specifically, through topological integration, the dynamic entity credit and capability profile set becomes a comprehensive data object that includes original probability inferences, derived credit indicators, classification capability labels, and their interrelationships, creating a digital profile of the entity. This profile not only provides concise indicators for decision-making (credit score, capability level) but also retains the underlying probabilistic basis and data relationships, enabling decision-makers to quickly grasp the entity's overview and delve into assessment details when needed, thus enhancing the transparency, auditability, and depth of decision support throughout the entire assessment process.

[0084] S3. Based on a dynamic entity credit and capability profile set, the flexible capacity portion of the total allowable nitrogen emission in the basin is allocated to each polluting entity through market mechanisms.

[0085] S3.1. Based on the watershed water environment model and water quality targets, calculate the total allowable discharge of total nitrogen in the watershed and divide the total allowable discharge of total nitrogen in the watershed into basic discharge capacity and flexible regulation capacity.

[0086] Furthermore, based on a pre-defined watershed water environment model and combined with inputs such as hydrological parameters, hydrodynamic conditions, and water quality standards for the target water area, numerical simulation is used to calculate the total allowable nitrogen discharge in the watershed under the constraint of water quality targets. This total amount represents the maximum environmental carrying capacity under specific protection objectives. The calculated total allowable nitrogen discharge is divided into two parts according to a pre-defined ratio or rule: a basic discharge capacity and a flexible adjustment capacity. The division rule can refer to the lower quantile of the historical discharge level of the polluting entity (such as the lowest discharge level within a certain period) or the unit product benchmark discharge quota based on industry standards. The sum of the total nitrogen emissions required to ensure that most polluting entities maintain basic, non-expansionary production activities is defined as the basic discharge capacity. The remaining portion of the total allowable nitrogen discharge minus the basic discharge capacity is defined as the flexible adjustment capacity. The basic discharge capacity is pre-allocated to each polluting entity in a relatively fixed manner to ensure the fairness of the basic right to survival and the right to development.

[0087] Specifically, transforming total environmental constraints into a dual structure combining a guaranteed baseline and competitive increments creates structural preconditions for differentiated and market-based allocation of environmental capacity. This division ensures, on the one hand, the bottom line of environmental justice (basic emission capacity) is guaranteed, and on the other hand, releases flexible space (flexible adjustment capacity) for optimized allocation, incentivizing advanced practices, and improving overall economic and environmental efficiency. This allows total emission control policies to balance the basic fairness goal of ensuring basic standards with the efficiency goal of promoting optimization.

[0088] S3.2 Based on the dynamic subject credit and capability profile set and evaluation criteria, construct a differentiated capacity application and evaluation mechanism for multiple rounds of sealed submission of allocation applications.

[0089] Furthermore, based on the dynamic entity credit and capability profile set, core quantitative evaluation indicators are extracted from the profile set, such as the comprehensive credit score and the classification label of governance capability level in the credit representation vector. Based on these core quantitative evaluation indicators, an evaluation priority coefficient is constructed through a pre-defined mapping function. The mapping function defines how to comprehensively transform the multi-dimensional evaluation information in the dynamic entity credit and capability profile set into a single, comparable influencing factor. For example, one mapping function can normalize the comprehensive credit score as the base coefficient, and then assign different multiplier factors according to the governance capability level (e.g., efficient, stable, basic compliance), and the final product is the evaluation priority coefficient. Specific rules for constructing a differentiated capacity application and evaluation mechanism include: specifying the application rounds, the time window for each round of application opening, the format and content of application information (e.g., the applicant's identity, the number of flexible adjustment capacities applied for), the method of application submission (sealed, i.e., information is not disclosed before the deadline), and the core allocation decision rule—that is, clearly announcing that the final allocation will be determined based on the application ranking adjusted by the evaluation priority coefficient.

[0090] Specifically, non-price attributes reflecting environmental performance (credit and capability) are systematically and transparently embedded into the market rules of resource allocation through evaluation priority coefficients. This changes the situation where price is the only competitive factor in the traditional market mechanism, creating a dual-core competitive dimension of price + environmental performance, or more precisely, demand expression + environmental performance coefficient. This avoids the environmental risks that may be caused by the highest bidder winning (high-pollution, high-profit industries may squeeze out industries with good environmental performance but low profits), and also avoids the inefficiency of allocation by pure administrative orders. It internalizes environmental management goals into the inherent screening criteria of the market mechanism, incentivizing polluters to commit to improving their environmental performance in the long term in order to gain a competitive advantage in the future, thus achieving an organic combination of environmental goals and market efficiency.

[0091] S3.3 In the constructed differentiated capacity application and evaluation mechanism, each polluting entity submits a sealed allocation application based on its demand for flexible adjustment capacity and the application priority determined by its own dynamic entity credit and capability profile.

[0092] Furthermore, under the publicly available rules of the differentiated capacity application and evaluation mechanism, each polluting entity operates as an independent decision-making unit. Each polluting entity first assesses its future demand for flexible adjustment capacity. This demand assessment is based on factors such as production plan forecasts, technological transformation progress, and changes in market orders. Each polluting entity can access its current dynamic entity credit and capability profile and, according to the published rules, calculate or obtain its evaluation priority coefficient within the current allocation cycle. In a sealed allocation application, the polluting entity must anonymously or encryptedly submit its identity and the quantity of flexible adjustment capacity it wishes to obtain, and sign the application with its digital certificate to ensure authenticity and non-repudiation. However, the specific value is not filled in by the polluting entity itself at the time of application, but is obtained by the mechanism operator from the background database based on the applicant's identity.

[0093] Specifically, when making decisions, the polluting entities are aware of the rules governing the impact of evaluation priority coefficients on ranking. Therefore, they will rationally weigh the options: given a certain demand, it is easier to obtain the required capacity, ensuring the implementation of the mechanism. This transforms abstract market rules and entity profiles into concrete and auditable participation behaviors. Through sealed submissions and coefficient correlation based on reliable backend data, the commercial privacy of each entity is protected (demand information is not disclosed in advance), while also eliminating the possibility of entities making false statements about their creditworthiness during the application process. This ensures the fairness and credibility of the entire competition process, allowing competition based on credit and capability to truly occur within a pure and credible framework.

[0094] S3.4 Based on the sealed allocation application results in the differentiated capacity application and evaluation mechanism, the flexible adjustment capacity is allocated to the top-ranked sewage discharge entities according to the application ranking adjusted by the evaluation priority coefficient.

[0095] Furthermore, after the application deadline stipulated by the differentiated capacity application and evaluation mechanism, the mechanism operator decrypts or summarizes all sealed allocation applications submitted by all polluting entities, obtaining the polluting entity identifier and the amount of flexible adjustment capacity applied for each application. Based on the polluting entity identifier, the evaluation priority coefficient corresponding to each applicant entity in the current cycle is retrieved from the background database. Then, according to the pre-published sorting algorithm, each application is calculated to generate a sorting basis. A typical sorting algorithm is to multiply the amount of flexible adjustment capacity applied for in each application by the evaluation priority coefficient of the applicant entity to obtain a priority-adjusted demand index. All applications are sorted from high to low according to the obtained priority-adjusted demand index. Under the constraint of a fixed total amount of flexible adjustment capacity, the flexible adjustment capacity of each application is satisfied in turn according to the above sorting order until all available flexible adjustment capacity is allocated, or all applications are satisfied (possibly incompletely). For applications that are not fully satisfied due to insufficient capacity, they can be recorded as partially satisfied or enter the next round of applications. Finally, a clear list of flexible regulation capacity allocation results is generated, specifying each sewage discharge entity that has been allocated capacity and the specific amount of flexible regulation capacity allocated to it.

[0096] Specifically, the quantity of demand (representing the intensity of economic willingness) and the evaluation priority coefficient (representing the quality of environmental performance) are integrated into a decision indicator through multiplication (or other mathematical combinations). Even if an entity has a large demand, its overall ranking may be low if its environmental performance (evaluation priority coefficient) is low. Conversely, entities with high environmental performance can occupy a favorable position in the ranking even if their individual demand is not large, thereby obtaining development resources commensurate with their environmental performance. This ensures that the flexible adjustment capacity can be prioritized for entities that both need development and are good at environmental protection, achieving a perfect balance between the economic allocation of environmental resources and the positive incentive of environmental performance. This allows the market mechanism to truly serve the dual goals of watershed environmental protection and high-quality development.

[0097] S4. Draft a differentiated load allocation scheme based on the basic capacity.

[0098] S4.1. The results of the flexible adjustment capacity allocation are summed and aggregated with the pre-set basic emission capacity to generate the final quota of basic emission capacity and the allocated flexible adjustment capacity for each pollutant, forming a draft of the differentiated load allocation scheme.

[0099] Furthermore, the basic emission capacity value of each pollutant is extracted from the completed basic capacity verification list, and the allocation amount corresponding to each pollutant is extracted from the flexible adjustment capacity allocation result list. The basic emission capacity value and the allocated flexible adjustment capacity value are added together to generate a complete data list containing the pollutant identification, basic emission capacity, allocated flexible adjustment capacity and final quota. This list constitutes the draft of the differentiated load allocation scheme.

[0100] Specifically, the system combines basic rights that guarantee rights with competitive incentives into a unified and enforceable individualized emission cap. Through simple arithmetic addition, it achieves a structural integration of the fair baseline and efficiency increments in the final management certificate. The base-plus quota composition model is solidified into different digital quotas for each entity, ensuring that incentives based on credit and capability can be directly and transparently transformed into actually usable emission space. For example, an entity with a high credit rating will have a larger final quota due to the additional flexible capacity it gains than an entity with a low credit rating, thereby transforming the dynamically assessed market competition results into a differentiated load allocation scheme.

[0101] S5. Associate and map the allocation results in the draft differentiated load allocation scheme with the assessment information in the dynamic subject credit and capability profile set to generate differentiated smart contract execution parameters.

[0102] S5.1 Based on the draft of the differentiated load allocation scheme, extract the final quota value allocated to each pollutant in the draft of the differentiated load allocation scheme to form a final quota allocation list of the final quota values.

[0103] Furthermore, the data structure of the draft differentiated load allocation scheme is analyzed, and the final quota value recorded for each pollutant discharge entity is read one by one. The unique identifier of each pollutant discharge entity is combined with its corresponding final quota value to generate a quota allocation record. The quota allocation records of all pollutant discharge entities within the watershed are then aggregated, arranged and stored in a unified format to form a structured final quota allocation list.

[0104] Specifically, the draft of the differentiated load allocation scheme was refined, transformed, and solidified into a concise, accurate, and machine-readable quota authorization data set. This enabled the lossless transfer of management decision results to the automated execution interface, ensuring accurate mapping from macro-level schemes to micro-level individual quota data. This provided an immutable and authoritative data foundation for the precise initialization of blockchain smart contracts.

[0105] S5.2 Based on the dynamic subject credit and capability profile set, extract the credit representation vector and governance capability level corresponding to each polluting subject in the dynamic subject credit and capability profile set to form a credit and capability attribute set of credit representation vector and governance capability level.

[0106] Furthermore, the system accesses the data records of each polluting entity stored in the dynamic entity credit and capability profile. From each polluting entity's data record, predefined credit representation vector fields (such as a multi-dimensional numerical vector representing comprehensive credit score, risk preference, etc.) and governance capability level fields (such as a classification label representing efficiency, stability, etc.) are extracted. The extracted credit representation vector and governance capability level are then associated and bound with the unique identifier of the polluting entity, forming an independent set of credit and capability attributes for each polluting entity. This completes the extraction of a subset of core attributes directly related to smart contract behavior control from a comprehensive and structured entity profile.

[0107] Specifically, the complex evaluation results, which reflect the subject's internal state and are obtained based on Bayesian inference and data feature analysis, underwent a key feature selection and interface processing to screen out the credit and capability dimensions most suitable for programmable and rule-based control, thereby creating conditions for transforming abstract evaluation conclusions into programmable contract logic.

[0108] S5.3. Based on the preset parameter mapping rules, the credit representation vector and the credit and capability attribute set of the governance capability level are mapped to smart contract function parameters.

[0109] Furthermore, based on predefined parameter mapping rules stored in the rule base, each attribute extracted from the credit and capability attribute set is processed. The parameter mapping rules define how to map credit and capability attribute values. For example, one parameter mapping rule might stipulate that the comprehensive credit score in the credit representation vector is mapped to the transaction fee discount rate parameter in the smart contract, with a higher score resulting in a larger discount rate. Another rule might stipulate that the governance capability level is efficiently mapped to the automatic write-off trigger threshold parameter of the smart contract, set to a more lenient threshold to encourage stable operation. By applying all relevant parameter mapping rules, the credit representation vector and governance capability level of each polluting entity are systematically converted into a series of specific smart contract functional parameter values ​​with clear business meanings, realizing semantic transformation and rule integration from the field of environmental credit and capability assessment to the field of smart contract programming and execution.

[0110] Specifically, a configurable, interpretable translation layer was created to translate the dynamically changing, multi-dimensional subject profile into control switches and numerical parameters that determine the specific execution behavior of smart contracts (such as fees, frequency, thresholds, and permissions). This allows smart contracts of reputable and capable polluting entities to automatically obtain more favorable, convenient, and flexible execution strategies in actual operation, thereby seamlessly and automatically embedding non-price incentives into the execution process through technical means.

[0111] S5.4 Based on the final quota allocation list and smart contract function parameters, generate a smart contract parameter configuration file for each polluting entity, aggregate the smart contract parameter configuration files of the polluting entities, and form a differentiated set of smart contract execution parameters.

[0112] Furthermore, the final quota value corresponding to the pollution discharge entity in the final quota allocation list is written into the total quota field of the pollution discharge entity's smart contract parameter configuration file. The set of smart contract function parameters (a series of parameter name and parameter value pairs) generated by the pollution discharge entity is written into the pollution discharge entity's smart contract parameter configuration file. A complete smart contract parameter configuration file containing its personalized final quota and differentiated function parameters is generated for each pollution discharge entity in the basin. The smart contract parameter configuration files of all pollution discharge entities are collected and packaged to form a complete set containing the differentiated configuration information of all entities, which constitutes the differentiated smart contract execution parameter set.

[0113] Specifically, two data streams from different sources (one being the management allocation result, and the other being the dynamic evaluation profile) and of different natures are perfectly integrated into a unified, object-oriented data structure. Each entity's smart contract parameter configuration file is an independent and complete digital contract blueprint, specifying both the amount that can be allocated (quota) and how it will be executed and monitored (credit-based parameters). This integration ensures a one-to-one correspondence and synchronous generation of differentiated allocation schemes and differentiated execution strategies at the technical implementation level, laying a solid data foundation for building a truly personalized, adaptive, and executable smart contract instance cluster.

[0114] S6. Differentiated smart contract execution parameters are deployed to the blockchain network to form a set of smart contracts with differentiated parameters.

[0115] S6.1. Based on the differentiated set of smart contract execution parameters, customize smart contract logic for each polluting entity and generate an individual smart contract deployment package.

[0116] Furthermore, using a predefined smart contract template, the system reads the smart contract parameter configuration file for a specific polluter from the differentiated smart contract execution parameter set. The final quota value in the smart contract parameter configuration file is assigned to the state variable representing the total quota in the smart contract template. All smart contract function parameters in the configuration file are assigned one by one to the corresponding control parameter variables or function modifiers in the smart contract template according to a preset mapping relationship. Through this assignment operation, the logical branches, execution conditions, and behavioral details of the general smart contract template are customized by specific parameter values, thereby generating a smart contract source code tailored to the polluter, containing specific quotas and differentiated rules. This source code is then compiled into bytecode executable by the blockchain virtual machine and encapsulated into an individual smart contract deployment package containing the bytecode, contract initialization data, and the polluter's identifier.

[0117] Specifically, the differentiated results of management decisions and entity assessments are transformed into personalized digital contracts that can be automatically executed on the blockchain. This is achieved through a templated parameter injection approach, enabling scalable customization. A single set of general smart contract logic, by injecting differentiated parameters, can generate a massive number of contract instances with varying execution strategies. For example, for entities with high credit ratings, the injected parameters will cause the contract to automatically execute lower transaction fees; for entities with strong governance capabilities, the injected parameters may allow for higher single-transaction write-off limits. This allows complex, differentiated governance rules to be transformed into automatically executable code, avoiding the tediousness and potential inconsistencies of writing separate contracts for each entity.

[0118] S6.2 Submit the individual smart contract deployment package to the permissioned consortium blockchain network. After network consensus verification, create and activate an independent smart contract account instance for each polluting entity on the blockchain.

[0119] Furthermore, individual smart contract deployment packages are submitted as transactions to a pre-established permissioned consortium blockchain network via a blockchain network interface. This network is jointly maintained by regulatory agencies, pollution entities, and third-party service providers as consensus nodes. The submitted deployment transactions are broadcast to all nodes in the network. The consensus nodes in the network verify the format compliance of the deployment transactions, the submitter's permissions, and the security of the smart contract bytecode according to a predetermined consensus algorithm, such as using a Byzantine fault-tolerant algorithm.

[0120] Specifically, deployment transactions verified through consensus are sorted and packaged into a new block. After the block is confirmed by the network, the smart contract bytecode contained in the deployment transactions in the block is permanently stored on the blockchain. A smart contract account instance with a unique address is then created, including information such as quota and functional parameters, thus entering a ready state that can be called externally. The customized smart contract logic is deployed to a trusted execution environment in a decentralized and tamper-proof manner. By utilizing the consensus mechanism and immutability of the blockchain, the deployment process and initial state of the differentiated allocation scheme and execution rules are made public, transparent, and trustworthy after being converted into code. No node can unilaterally tamper with the deployed contract or its initial parameters, which lays a technological trust foundation for the fairness and authority of subsequent automated execution.

[0121] S6.3. Aggregate smart contract account instances created and activated on the blockchain to form a set of smart contracts with differentiated parameters.

[0122] Furthermore, on the permissioned consortium blockchain network, newly created and activated smart contract account instances are scanned periodically or in real-time. By querying the blockchain's state or event logs, smart contract account instances created by specific deployment transactions and associated with the pollution entity's identifier are identified. The unique contract addresses, corresponding pollution entity identifiers, and initialization state summaries of all such activated smart contract account instances are collected and organized into an index or list reflecting the ready-to-execute smart contract entities corresponding to each pollution entity on the blockchain. By aggregating the smart contract account instance information of all pollution entities, a set of smart contracts with differentiated parameters is formed, providing final confirmation and integration of the blockchain deployment results.

[0123] Specifically, smart contract instances are logically aggregated into a unified, globally viewable contract cluster. This cluster represents the final technical manifestation of all preceding differentiated decisions and parameter configurations, signifying a collection of digitized, personalized, and ready-to-execute contracts on the blockchain for all polluting entities within the watershed. Managers can use this cluster to gain a macro-level understanding of the deployment status of all contracts. Subsequent automated execution processes are triggered by calling specific contract instance addresses within this cluster, marking the formation of a complete closed loop from management solutions to a trusted, automated execution platform.

[0124] S7. Obtain new multi-source standardized datasets through off-chain trusted data sources, trigger the automatic execution of quota management of a set of smart contracts with differentiated parameters on the blockchain network, and generate on-chain trusted execution records.

[0125] S7.1 The off-chain trusted data source collects and processes data from the online monitoring system of the polluting enterprise according to a preset cycle, and generates the latest multi-source standardized dataset.

[0126] Furthermore, one or more off-chain trusted data sources are pre-configured and authorized. These off-chain trusted data sources have the authority to collect data from the online monitoring systems of polluting enterprises, environmental monitoring points, and other certified data providers. The off-chain trusted data sources automatically initiate data collection tasks according to fixed time periods set by management requirements, such as daily or weekly, accessing and obtaining raw information such as real-time or near-real-time emission monitoring data and treatment facility operation logs from each polluting entity. The acquired multi-source heterogeneous data is sent to a data processing flow. This flow performs similar spatiotemporal alignment and data cleaning operations for incremental data, ensuring that the new data conforms to specifications in terms of timestamps, coordinates, and quality. The processed data is organized into a multi-source standardized dataset consistent with historical data formats, serving as the latest evidence set reflecting the actual performance of the polluting entities in the most recent time period.

[0127] S7.2 The latest multi-source standardized dataset is signed and transmitted to the blockchain network in the form of oracle transactions. The set of smart contracts with differentiated parameters on the blockchain network receives the oracle transactions, automatically parses the emission data in the latest multi-source standardized dataset, and executes quota verification, transaction and reward / penalty operations according to the built-in logic of the contract to generate a trusted execution record on the chain.

[0128] Furthermore, the off-chain trusted data source uses its private key to digitally sign the newly generated multi-source standardized dataset to prove the authenticity and integrity of the data source. The signed latest multi-source standardized dataset, along with the signature itself and necessary metadata such as timestamps, is assembled into a structured oracle transaction. This transaction is deployed within a set of parameter-differentiated smart contracts on the blockchain, pre-configured with logic to listen for events at specific oracle addresses. When the oracle transaction associated with the parameter-differentiated smart contract set is confirmed and uploaded to the chain, the transaction content is captured as an event by the relevant smart contract account instance. The smart contract account instance automatically calls the built-in parsing function to extract the emission data of the pollution entity corresponding to the contract from the latest multi-source standardized dataset in the oracle transaction. Based on the final quota value injected during initialization, the smart contract function parameters, and the pre-defined business logic, all the details, results, and timestamps of these operations are permanently recorded on the blockchain as transaction logs and state changes, forming an immutable and traceable on-chain trusted execution record.

[0129] Specifically, a set of smart contracts with differentiated parameters can automatically execute based on the differentiated rules of each contract instance after receiving trusted data, without human intervention. This ensures that the process from data to action is objective and tamper-proof. For example, a contract of a creditworthy entity may automatically complete the write-off and award points upon receiving compliant data; while a contract of a creditworthy entity may only complete the write-off without awarding points when receiving the same compliant data, or even trigger more severe automatic penalties immediately when the data is abnormal, generating a legally valid and trusted execution trajectory.

[0130] S8. Update the dynamic entity credit and capability profile set by combining the on-chain trusted execution records with the new multi-source standardized dataset, and adjust the set of smart contracts with differentiated parameters.

[0131] S8.1. The on-chain trusted execution records and the new multi-source standardized dataset are used as a new evidence stream and input into the Bayesian update engine. Based on the new evidence stream, the original belief state probability distribution is iteratively updated to generate an updated dynamic subject credit and capability profile set.

[0132] Furthermore, trusted on-chain execution records automatically generated by a set of smart contracts with differentiated parameters are obtained from the blockchain, such as quota verification records, transaction records, and reward / penalty trigger records. Simultaneously, the latest multi-source standardized dataset generated concurrently from trusted off-chain data sources is acquired. The trusted on-chain execution records and the latest multi-source standardized dataset are timestamped and correlated, merging to form a new evidence stream. This new evidence stream is input into the previously constructed Bayesian update engine. Based on a pre-defined cognitive framework of the polluting entity's behavior, the Bayesian update engine analyzes the correlation strength between the new evidence stream and each hidden state hypothesis, obtaining a new multi-dimensional likelihood correlation strength spectrum.

[0133] Specifically, the Bayesian update engine uses the new multi-dimensional likelihood correlation strength spectrum as the likelihood function and the current belief state probability distribution, which reflects the previous round of cognition, as the prior probability to obtain the updated posterior probability distribution, i.e., the new belief state probability distribution. Based on this new belief state probability distribution, the credit representation mapping rule and the capability validity extraction rule are applied again, and the governance facility operation data in the latest multi-source standardized dataset are combined to recalculate the credit representation vector and governance capability level. The new belief state probability distribution, the re-acquired credit representation vector, and the governance capability level are topologically linked and integrated to form an updated dynamic subject credit and capability profile set.

[0134] S8.2 Based on the updated dynamic subject credit and capability profile set, the execution logic in the set of smart contracts with differentiated parameters on the blockchain network is dynamically reconfigured through parameter adjustment transactions, thereby adjusting the set of smart contracts with differentiated parameters.

[0135] Furthermore, an updated dynamic entity credit and capability profile set is obtained. From this profile, the updated credit representation vector and governance capability level of each polluting entity are extracted. Based on a preset parameter mapping rule, the updated credit representation vector and governance capability level are remapped to a new set of smart contract function parameters. A parameter adjustment transaction is generated for each polluting entity. This transaction includes the address of the target smart contract account instance, the name of the parameter to be updated, and the corresponding new parameter value. The generated parameter adjustment transaction is then submitted to the permissioned consortium blockchain network.

[0136] Specifically, after network consensus verification, the parameter adjustment transaction is confirmed and executed. The execution process involves calling the pre-defined parameter update function in the corresponding smart contract account instance, writing the new smart contract function parameters into the smart contract's state storage, overwriting or modifying the original parameter settings. The execution logic of each affected smart contract account instance in the set of differentiated smart contracts, such as transaction fee rates, automatic write-off thresholds, and reward / penalty coefficients, is dynamically reconfigured based on the updated dynamic entity credit and capability profile, thereby achieving the adjustment of the set of differentiated smart contracts.

[0137] This embodiment also provides a multi-entity total nitrogen emission load allocation system based on multi-source data, including: a processing module that collects and processes multi-source heterogeneous data and outputs a multi-source standardized dataset;

[0138] The inference module uses a Bayesian update engine to infer the hidden credit status and governance capacity level of each polluting entity, and outputs a dynamic entity credit and capacity profile set.

[0139] The allocation module, based on a dynamic entity credit and capability profile set, allocates the elastic capacity portion of the total allowable nitrogen emissions in the basin to each polluting entity through market mechanisms, and forms a draft differentiated load allocation scheme in combination with the basic capacity;

[0140] The mapping module associates and maps the allocation results in the draft of the differentiated load allocation scheme with the evaluation information in the dynamic subject credit and capability profile set, generates differentiated smart contract execution parameters, and deploys the differentiated smart contract execution parameters to the blockchain network to form a set of smart contracts with differentiated parameters.

[0141] The execution module obtains new multi-source standardized datasets through off-chain trusted data sources, triggers a set of smart contracts with differentiated parameters on the blockchain network to automatically execute quota management, and generates on-chain trusted execution records;

[0142] The update module updates the dynamic entity credit and capability profile set by combining on-chain trusted execution records with a new multi-source standardized dataset and adjusting the set of smart contracts with differentiated parameters.

[0143] This embodiment also provides a computer device applicable to the multi-entity total nitrogen emission load allocation method based on multi-source data, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the multi-entity total nitrogen emission load allocation method based on multi-source data as proposed in the above embodiment.

[0144] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0145] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the multi-entity total nitrogen emission load allocation method based on multi-source data as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0146] In summary, this invention aggregates and processes heterogeneous data from multiple sources to form a standardized dataset. It then uses a Bayesian update engine to dynamically infer the credit status and governance capacity of polluting entities, generating a dynamic profile of their credit and capabilities. Combining basic capacity with elastic capacity allocated through market mechanisms, a draft differentiated load allocation scheme is formed. The allocation results and evaluation information are mapped to differentiated smart contract execution parameters, which are then deployed to a blockchain network to form a set of smart contracts with differentiated parameters. The smart contracts are automatically executed for quota management via off-chain trusted data sources, generating on-chain trusted execution records. These records, along with new datasets, are used to update the profile and adjust the smart contracts. This achieves fully automated, precise, and traceable load allocation management from data collection, dynamic evaluation, and intelligent allocation to trusted execution and closed-loop feedback.

[0147] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-entity total nitrogen emission load allocation method based on multi-source data, characterized in that: This includes collecting and processing heterogeneous data from multiple sources, and outputting standardized datasets from multiple sources; Using a Bayesian update engine, the hidden credit status and governance capacity level of each polluting entity are inferred, and a dynamic entity credit and capacity profile is output, including the following steps: Based on a multi-source standardized dataset, a hidden state hypothesis space with multiple potential behavioral patterns is defined for each sewage discharge entity. Prior probability weights are configured for the hidden state hypothesis space according to industry benchmarks and historical records to form an initial state probability distribution and construct a Bayesian update engine. Using multi-source standardized datasets as input to the Bayesian update engine for dynamic evidence stream, within the cognitive framework of the pollutant subject's behavior, the correlation strength between the dynamic evidence stream and each hidden state is analyzed, and a multi-dimensional likelihood correlation strength spectrum is generated. The cognitive weight allocation of the initial state probability distribution is adjusted based on the multidimensional likelihood correlation strength spectrum, and the belief state probability distribution reflecting the degree of support of the latest evidence is evolved through iterative updates. Based on the aforementioned belief state probability distribution, the credit representation mapping rule is used to condense high-dimensional probabilistic beliefs into low-dimensional operable credit representation vectors. Based on the probability distribution of belief states, the robustness measure of the operation mode that oscillates in tandem with the high integrity state belief in the hidden state hypothesis space is analyzed from the governance facility operation data fragments of the multi-source standardized dataset using the capability validity extraction rule. The robustness measure of the operating mode is compared with the preset efficiency threshold spectrum to determine the treatment capacity level of each pollutant. The probability distribution of each pollutant's belief state, credit representation vector, and governance capacity level are topologically linked and integrated to form a dynamic profile of the pollutant's credit and capacity. Based on a dynamic entity credit and capability profile, the flexible capacity portion of the total allowable nitrogen emissions in the watershed is allocated to each polluting entity through market mechanisms, including the following steps: Based on the watershed water environment model and water quality targets, the total allowable discharge of total nitrogen in the watershed is calculated and divided into basic discharge capacity and flexible regulation capacity. Based on a dynamic entity credit and capability profile and evaluation criteria, a differentiated capacity application and evaluation mechanism is constructed for multiple rounds of sealed submission of allocation applications; In the differentiated capacity application and evaluation mechanism, each polluting entity submits a sealed allocation application based on its demand for flexible adjustment capacity and the application priority determined by its own dynamic entity credit and capability profile. Based on the sealed allocation application results in the differentiated capacity application and evaluation mechanism, the flexible adjustment capacity is allocated to the top-ranked sewage discharge entities according to the application ranking adjusted by the evaluation priority coefficient. A draft differentiated load allocation scheme is developed by combining the basic capacity, including the following steps: The results of the flexible adjustment capacity allocation are summed and aggregated with the pre-set basic emission capacity to generate the final quota of basic emission capacity and the allocated flexible adjustment capacity for each pollutant, forming a draft of the differentiated load allocation scheme. The allocation results in the draft differentiated load allocation scheme are correlated and mapped with the assessment information in the dynamic subject credit and capability profile set to generate differentiated smart contract execution parameters. The differentiated smart contract execution parameters are then deployed to the blockchain network to form a set of smart contracts with differentiated parameters. By acquiring new multi-source standardized datasets through off-chain trusted data sources, a set of smart contracts with differentiated parameters on the blockchain network is triggered to automatically execute quota management and generate on-chain trusted execution records. By combining on-chain trusted execution records with new multi-source standardized datasets, we update the dynamic entity credit and capability profile set and adjust the set of smart contracts with differentiated parameters.

2. The multi-entity total nitrogen emission load allocation method based on multi-source data as described in claim 1, characterized in that: Collecting and processing heterogeneous data from multiple sources to output a standardized multi-source dataset includes the following steps: Based on the spatiotemporal alignment of the collected multi-source heterogeneous data, a spatiotemporal aligned data set is generated. The spatiotemporal aligned data set is then cleaned to obtain a multi-source standardized dataset.

3. The multi-entity total nitrogen emission load allocation method based on multi-source data as described in claim 1, characterized in that: The allocation results in the draft differentiated load allocation scheme are correlated and mapped with the assessment information in the dynamic entity credit and capability profile set to generate differentiated smart contract execution parameters, including the following steps: Based on the draft of the differentiated load allocation scheme, the final quota value allocated to each pollutant in the draft of the differentiated load allocation scheme is extracted to form a final quota allocation list. Based on the dynamic subject credit and capability profile set, the credit representation vector and governance capability level corresponding to each polluting subject in the dynamic subject credit and capability profile set are extracted to form a credit and capability attribute set of credit representation vector and governance capability level. Based on the preset parameter mapping rules, the credit representation vector and the credit and capability attribute set of governance capability level are mapped to smart contract function parameters. Based on the final quota allocation list and smart contract function parameters, a smart contract parameter configuration file is generated for each polluting entity. The smart contract parameter configuration files of the polluting entities are aggregated to form a differentiated set of smart contract execution parameters.

4. The multi-entity total nitrogen emission load allocation method based on multi-source data as described in claim 1, characterized in that: Differentiated smart contract execution parameters are deployed to the blockchain network to form a set of smart contracts with differentiated parameters, including the following steps: Based on a differentiated set of smart contract execution parameters, smart contract logic is customized for each polluting entity to generate an individual smart contract deployment package; Individual smart contract deployment packages are submitted to the permissioned consortium blockchain network. After verification by network consensus, an independent smart contract account instance is created and activated on the blockchain for each polluting entity. The smart contract account instances created and activated on the blockchain are aggregated to form a set of smart contracts with differentiated parameters.

5. The multi-entity total nitrogen emission load allocation method based on multi-source data as described in claim 1, characterized in that: By acquiring a new multi-source standardized dataset through trusted off-chain data sources, and triggering the automatic execution of a set of smart contracts with differentiated parameters on the blockchain network for quota management, a trusted on-chain execution record is generated, including the following steps: The off-chain trusted data source collects and processes data from the online monitoring system of polluting enterprises and other sources according to a preset cycle, generating the latest multi-source standardized dataset. The latest multi-source standardized dataset is signed and transmitted to the blockchain network in the form of oracle transactions. The set of smart contracts with differentiated parameters on the blockchain network receives the oracle transactions, automatically parses the emission data in the latest multi-source standardized dataset, and executes quota verification, transaction and reward / penalty operations according to the built-in logic of the contract, generating a trusted on-chain execution record.

6. The multi-entity total nitrogen emission load allocation method based on multi-source data as described in claim 5, characterized in that: The process involves updating the dynamic entity credit and capability profile set using on-chain trusted execution records and a new multi-source standardized dataset, and adjusting the set of smart contracts with differentiated parameters, including the following steps: The on-chain trusted execution records and the new multi-source standardized dataset are used as a new evidence stream and input into the Bayesian update engine. Based on the new evidence stream, the original belief state probability distribution is iteratively updated to generate an updated dynamic subject credit and capability profile set. Based on the updated dynamic entity credit and capability profile set, the execution logic of the set of smart contracts with differentiated parameters on the blockchain network is dynamically reconfigured through parameter adjustment transactions, thereby adjusting the set of smart contracts with differentiated parameters.

7. A multi-entity total nitrogen emission load allocation system based on multi-source data, based on the multi-entity total nitrogen emission load allocation method based on multi-source data as described in any one of claims 1 to 6, characterized in that: This includes a processing module that aggregates and processes heterogeneous data from multiple sources, and outputs a standardized dataset from multiple sources. The inference module uses a Bayesian update engine to infer the hidden credit status and governance capacity level of each polluting entity, and outputs a dynamic entity credit and capacity profile set. The allocation module, based on a dynamic entity credit and capability profile set, allocates the elastic capacity portion of the total allowable nitrogen emissions in the basin to each polluting entity through market mechanisms, and forms a draft differentiated load allocation scheme in combination with the basic capacity; The mapping module associates and maps the allocation results in the draft of the differentiated load allocation scheme with the evaluation information in the dynamic subject credit and capability profile set, generates differentiated smart contract execution parameters, and deploys the differentiated smart contract execution parameters to the blockchain network to form a set of smart contracts with differentiated parameters. The execution module obtains new multi-source standardized datasets through off-chain trusted data sources, triggers a set of smart contracts with differentiated parameters on the blockchain network to automatically execute quota management, and generates on-chain trusted execution records; The update module updates the dynamic entity credit and capability profile set by combining on-chain trusted execution records with a new multi-source standardized dataset and adjusting the set of smart contracts with differentiated parameters.

Images