Sludge resource utilization path determination method, device and equipment

Abstract

The invention provides a method, a device and equipment for determining a sludge resource utilization path, belongs to the technical field of computer information processing, and solves the problem that economic, environmental and social targets are difficult to dynamically and collaboratively optimize in the current sludge resource treatment process. The method comprises the following steps: acquiring original data, wherein the original data comprises at least one of sludge attribute data, process parameters and external dynamic data; preprocessing the original data to obtain sludge sample data; classifying and integrating the sludge sample data to obtain a multi-dimensional tensor; decomposing and dynamically updating the multi-dimensional tensor to obtain a decomposed factor matrix; inputting the decomposed factor matrix into a path optimization model for processing to obtain a resource optimization path set; and performing optimal verification processing on the resource optimization path set to obtain optimal utilization path data. According to the scheme, dynamic optimal balance of sludge treatment economic benefits, carbon emission reduction and social compliance is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of computer information processing, and in particular to a method, device and equipment for determining a sludge resource utilization path. Background Art

[0002] With the acceleration of urbanization, sludge production continues to rise, and its resource utilization has become a key issue in the environmental protection field. Traditional optimization methods often focus on a single economic or environmental objective, using static weight allocation or linear programming solutions. These methods are difficult to adapt to complex scenarios involving dynamic external environments (such as fluctuating energy prices) and conflicting multiple objectives.

[0003] When processing multi-source heterogeneous data, existing technologies often lead to the loss of key information due to dimensional collapse or modal splitting. For example, the synergistic effect between process parameters and policy texts has not been effectively modeled. In addition, most models rely on offline training of historical data and lack real-time feedback and dynamic update mechanisms, which can easily lead to decision failures due to data distribution offsets or sudden changes in the environment. It is particularly worth noting that existing methods are insufficient in quantitative evaluation of policies, making it difficult to map abstract policy clauses into computable constraints, resulting in resource utilization paths facing the risk of violations. The above-mentioned defects restrict the sustainability and actual implementation effectiveness of sludge resource utilization decisions, and urgently need to achieve technological breakthroughs through cross-modal data fusion, dynamic game optimization and closed-loop feedback mechanisms. Summary of the Invention

[0004] The present invention provides a method, device and equipment for determining a sludge resource utilization path, which solves the problem that economic, environmental and social goals are difficult to dynamically coordinate and optimize in the current sludge resource utilization process.

[0005] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0006] An embodiment of the present invention provides a method for determining a sludge resource utilization path, comprising:

[0007] Acquiring raw data, wherein the raw data includes at least one of sludge attribute data, process parameters, and external dynamic data;

[0008] Preprocessing the raw data to obtain sludge sample data;

[0009] Classifying and integrating the sludge sample data to obtain a multidimensional tensor;

[0010] Decomposing and dynamically updating the multidimensional tensor to obtain a decomposed factor matrix;

[0011] Inputting the decomposed factor matrix into a path optimization model for processing to obtain a resource optimization path set;

[0012] An optimal verification process is performed on the resource optimization path set to obtain optimal utilization path data.

[0013] Optionally, obtaining the original data includes:

[0014] Acquiring the sludge property data through a sensor array or laboratory testing, wherein the sludge property data includes at least one of organic matter content, heavy metal concentration, and moisture content;

[0015] Obtaining the process parameters through real-time recording of the sludge treatment equipment monitoring system, wherein the process parameters include at least one of reaction temperature, pH value, residence time, and unit energy consumption;

[0016] Obtaining numerical data from the external dynamic data through a preset database or a preset platform;

[0017] Obtaining text data from the external dynamic data through preset text information;

[0018] The spatiotemporal data in the external dynamic data is obtained by associating the geographic information system with the timestamp tag.

[0019] Optionally, the raw data is preprocessed to obtain sludge sample data, including:

[0020] Normalizing the sludge attribute data, the process parameters, and the numerical data to obtain normalized data;

[0021] Performing vectorization processing on the text data to obtain a semantic vector;

[0022] Performing missing filling processing on the spatiotemporal data to obtain filled spatiotemporal data;

[0023] The normalized data, the semantic vector and the filled spatiotemporal data are aligned according to a preset dimension to obtain sludge sample data.

[0024] Optionally, the sludge sample data is classified and integrated to obtain a multidimensional tensor, including:

[0025] Classifying the sludge sample data according to preset dimensions to obtain a plurality of classified data; wherein the preset dimensions include at least one of a sample dimension, a feature dimension, an external factor dimension, a time dimension, and a space dimension;

[0026] Integrating the multiple classification data to obtain a multidimensional tensor, wherein the multidimensional tensor is represented as: T(i, j, k, l, m);

[0027] Where i, j, k, l, and m are natural numbers, and T(i, j, k, l, m) is the observed value of the jth feature of the i-th sample under the influence of the k-th external factor, in the l-th time period and the m-th region.

[0028] Optionally, the multidimensional tensor is decomposed and dynamically updated to obtain a decomposed factor matrix, including:

[0029] Decomposing the multidimensional tensor data to obtain a set of low-rank factor matrices;

[0030] The time dimension factor matrix in the low-rank factor matrix set is dynamically updated to obtain a decomposed factor matrix.

[0031] Optionally, the path optimization model includes at least one of an economic agent, an environmental agent, and a social agent;

[0032] The state space of the economic agent includes at least one of real-time energy price, sludge calorific value and tensor decomposition eigenvector; the action space of the economic agent includes at least one of pyrolysis action, composting action and incineration action; the reward function of the economic agent is: R E =E p -α·(E h +T c ), α is the cost penalty coefficient, R E is the reward value of the economic agent, E p is the expected profit, E h is the expected processing cost, T c For transportation costs;

[0033] The state space of the environmental agent includes at least one of real-time carbon emission intensity, carbon tax unit price and historical carbon emission average; the action space of the environmental agent includes carbon emission constraint threshold; the reward function of the environmental agent is: R Env =-β·(C e ×C t ), β is the environmental penalty coefficient, R Env is the reward value of the environment agent, C e is carbon emissions, C t is the unit price of carbon tax;

[0034] The state space of the social agent includes at least one of a semantic vector and a number of historical violations; the action space of the social agent includes a compliance score; and the reward function of the social agent is: R S =γ·C s , γ is the compliance reward coefficient, R S is the reward value of the social agent, C s Score compliance.

[0035] Optionally, performing optimal verification processing on the resource optimization path set to obtain optimal utilization path data includes:

[0036] Extracting the resource optimization path set to obtain a first candidate solution and a second candidate solution;

[0037] Determine the dominance relationship between the first candidate solution and the second candidate solution according to a preset dominance condition to determine a dominated solution set and a non-dominated solution set;

[0038] The non-dominated solution set is sorted according to the preset target priority to obtain the optimal utilization path data.

[0039] Optionally, the method for determining a sludge resource utilization path further includes:

[0040] Obtain actual processing effect data;

[0041] Inputting the actual processing effect data and the optimal utilization path data into a loss function for processing to obtain deviation data;

[0042] performing an updating process on the decomposed factor matrix according to the deviation data to obtain an updated factor matrix;

[0043] The path optimization model is updated according to the updated factor matrix to obtain an updated path optimization model.

[0044] An embodiment of the present invention further provides a device for determining a sludge resource utilization path, comprising:

[0045] An acquisition module, configured to acquire raw data, wherein the raw data includes at least one of sludge attribute data, process parameters, and external dynamic data;

[0046] a processing module configured to pre-process the raw data to obtain sludge sample data; classify and integrate the sludge sample data to obtain a multidimensional tensor; decompose and dynamically update the multidimensional tensor to obtain a decomposed factor matrix; and input the decomposed factor matrix into a path optimization model for processing to obtain a resource optimization path set;

[0047] The determination module is used to perform optimal verification processing on the resource optimization path set to obtain optimal utilization path data.

[0048] An embodiment of the present invention further provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program executes the above method when executed by the processor.

[0049] The technical solution of the present invention includes at least the following effects:

[0050] The above-mentioned scheme of the present invention obtains original data, which includes at least one of sludge attribute data, process parameters and external dynamic data; pre-processes the original data to obtain sludge sample data; classifies and integrates the sludge sample data to obtain a multidimensional tensor; decomposes and dynamically updates the multidimensional tensor to obtain a decomposed factor matrix; inputs the decomposed factor matrix into the path optimization model for processing to obtain a resource optimization path set; performs optimal verification processing on the resource optimization path set to obtain optimal utilization path data, thereby achieving a dynamic optimal balance among the economic benefits, carbon emission reduction and social compliance of sludge treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of a method for determining a sludge resource utilization path provided by an embodiment of the present invention;

[0052] Figure 2 This is a structural diagram of a device for determining a sludge resource utilization path provided by an embodiment of the present invention;

[0053] Figure 3 It is a structural diagram of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0055] like Figure 1 As shown, an embodiment of the present invention proposes a method for determining a sludge resource utilization path, comprising:

[0056] Step 11, obtaining raw data, wherein the raw data includes at least one of sludge attribute data, process parameters, and external dynamic data;

[0057] Step 12: pre-processing the raw data to obtain sludge sample data;

[0058] Step 13: classify and integrate the sludge sample data to obtain a multidimensional tensor;

[0059] Step 14: Decomposing and dynamically updating the multidimensional tensor to obtain a decomposed factor matrix;

[0060] Step 15: Input the decomposed factor matrix into the path optimization model for processing to obtain a resource optimization path set;

[0061] Step 16: Perform optimal verification processing on the resource optimization path set to obtain optimal utilization path data.

[0062] In this embodiment, the raw data includes at least one of sludge attribute data, process parameters, and external dynamic data. Sludge attribute data includes the physical, chemical, and biological properties of the sludge. This data helps determine appropriate treatment and resource utilization methods. For example, sludge with a high moisture content may require dehydration, while sludge with excessively high heavy metal content requires special treatment processes to prevent environmental pollution. Process parameters involve operating parameters at various stages of the sludge treatment process, such as temperature, pressure, reaction time, stirring speed, and reagent dosage. Different process parameters can significantly affect the sludge treatment effect and resource utilization efficiency. For example, during the pyrolysis process of sludge, controlling temperature and reaction time directly affects the quality and yield of the pyrolysis products. External dynamic data includes market price information, regulatory requirements, and environmental factors. These external factors adjust dynamically over time and environmental changes, significantly impacting the feasibility and economic viability of sludge resource utilization. For example, rising market prices for sludge treatment products may prompt companies to adopt more efficient resource utilization processes to increase production. Raw data can be obtained through various channels, such as on-site sampling and testing, online monitoring equipment, laboratory analysis, market research, and policy document collection, ensuring the accuracy, completeness, and timeliness of the data.

[0063] After obtaining the raw data, it needs to be preprocessed to obtain the sludge sample data. The preprocessing process includes: data cleaning, which removes noise, outliers, and duplicate data from the raw data; data standardization, which eliminates the impact of dimensions because different types of data have different dimensions and value ranges; data missing value processing, which can cause deviations in data analysis results. Therefore, it is necessary to select an appropriate processing method based on the actual data situation; and data conversion, which converts the data appropriately according to the needs of subsequent analysis.

[0064] The process of classifying and integrating sludge sample data involves data classification and data integration. Data classification involves categorizing sludge sample data based on its characteristics and intended use. For example, sludge can be classified by source (e.g., domestic sludge, industrial sludge), treatment stage (e.g., pretreatment, advanced treatment), or resource utilization (e.g., fertilizer production, brick production, power generation). The purpose of classification is to better organize and manage data for subsequent analysis and processing. Data integration involves integrating the classified data to construct a multidimensional data structure. Multidimensional tensors are an effective mathematical tool for representing multidimensional data and can simultaneously incorporate information from multiple dimensions. For example, a three-dimensional tensor can be constructed, where the three dimensions represent the sludge source, treatment process, and resource utilization product, respectively. Each element corresponds to relevant data on the resource utilization product obtained under that specific source and treatment process, such as yield and quality indicators. By integrating sludge sample data into a multidimensional tensor, a more comprehensive description of the various factors and their interrelationships in the sludge resource utilization process can be achieved, providing richer information for subsequent decomposition and dynamic update processing.

[0065] After determining the multidimensional tensor, an appropriate tensor decomposition algorithm is used to decompose the multidimensional tensor. The purpose of tensor decomposition is to decompose the complex multidimensional tensor into the product form of multiple low-dimensional factor matrices, each factor matrix corresponding to a dimension of the tensor. These factor matrices contain the characteristic information of the original data in different dimensions, which helps to deeply understand the key factors and influencing factors in the sludge resource utilization process. Since the sludge resource utilization process is a dynamic process, the original data will continue to change over time. Therefore, it is necessary to dynamically update the decomposed factor matrix to reflect the latest changes in the data. The dynamic update method can be selected according to the actual situation to ensure that the treatment results are always consistent with the actual situation and improve the accuracy and reliability of the sludge resource utilization path.

[0066] According to the objectives and constraints of sludge resource utilization, a suitable path optimization model is constructed. Common optimization objectives include maximizing resource utilization efficiency, minimizing treatment costs, and reducing environmental pollution. Constraints include process technology feasibility, equipment capacity limitations, environmental protection regulations, etc. The path optimization model can be constructed using mathematical programming methods. Specifically, the decomposed factor matrix is ​​used as input data and substituted into the path optimization model for solution. Through iterative calculations of the optimization algorithm, the optimal resource utilization path set that meets the optimization objectives and constraints is found. For example, considering the goals of maximizing resource utilization efficiency and minimizing treatment costs, the optimization model may recommend a variety of different treatment process combinations and resource utilization plans to form a resource optimization path set. These path sets contain optimal or suboptimal solutions under different conditions, providing a scientific basis for decision-making on sludge resource utilization.

[0067] To identify the optimal utilization path from a set of resource optimization paths, appropriate validation metrics must be selected. These metrics should be relevant to the optimization objective and comprehensively reflect the strengths and weaknesses of the resource utilization path. For example, the yield, quality, economic benefits, and environmental benefits of the resource-based product can be selected as validation metrics. Each path in the set of resource optimization paths can be validated using experimental verification, simulation, and case studies. Experimental verification can assess the feasibility and effectiveness of a path through actual pilot, pilot, or production trials. Simulation verification can use computer simulation software to simulate the path and analyze its performance metrics. Case studies can compare the actual performance of different paths by referencing existing similar project examples. By combining multiple validation methods, the validation metrics of each path are evaluated and compared. Based on the validation results, the path that performs best in terms of these metrics is selected as the optimal utilization path. Relevant data on the optimal utilization path, such as treatment process parameters, resource-based product information, and economic and environmental benefits, should be organized and recorded to form optimal utilization path data. This data can provide detailed guidance for the implementation of sludge resource utilization projects, ensuring that they achieve their intended goals and outcomes.

[0068] The above-mentioned embodiments of the present invention construct a spatiotemporal correlation tensor by classifying and preprocessing multi-source data and fusing features; extract potential features based on sparse non-negative decomposition and dynamically update the weights of external factors; design economic, environmental, and social intelligent agents to drive path selection, carbon emission constraints and compliance scores respectively, and generate a frontier solution set through a balance strategy; finally, use actual processing effect data to construct a multi-objective feedback loss function, synchronously update tensor model parameters and decision-making strategies, and form a "data-model-decision" closed-loop adaptive optimization system. Combined with the multi-agent reinforcement learning collaborative game mechanism, it realizes multi-objective dynamic optimization of economic profits, carbon emissions and policy compliance, and improves the robustness and sustainability of resource utilization paths in complex dynamic environments.

[0069] In an optional embodiment of the present invention, step 11 may include:

[0070] Step 111, obtaining the sludge property data through a sensor array or laboratory testing, wherein the sludge property data includes at least one of organic matter content, heavy metal concentration, and moisture content;

[0071] Step 112, obtaining the process parameters through real-time recording of the sludge treatment equipment monitoring system, wherein the process parameters include at least one of reaction temperature, pH value, residence time, and specific energy consumption;

[0072] Step 113: obtaining numerical data from the external dynamic data through a preset database or a preset platform;

[0073] Step 114, obtaining text data in the external dynamic data through preset text information;

[0074] Step 115 : obtaining the spatiotemporal data in the external dynamic data by associating it with the timestamp tag through a geographic information system.

[0075] In this embodiment, step 111 involves deploying multiple sensors at the sludge treatment site to monitor key physical and chemical indicators of the sludge in real time, such as organic matter content, heavy metal concentration, and moisture content. Sensor data is transmitted to a data processing center via wireless or wired means. For indicators that cannot be monitored in real time by sensors, or for those requiring higher-precision measurements, sludge samples can be collected regularly and sent to a laboratory for testing. Laboratory test results serve as a supplement or verification of the sensor data.

[0076] In step 112, the monitoring system equipped with the sludge treatment equipment records key process parameters during the treatment process, such as reaction temperature, pH, residence time, and specific energy consumption, in real time. The monitoring system transmits these real-time process parameters to the data processing center via sensors and data acquisition modules. The data collection frequency can be adjusted based on actual needs to ensure real-time and accurate data. The collected process parameters are verified to ensure data integrity and consistency. Abnormal data or missing values ​​are marked and processed to prevent them from affecting subsequent data analysis and optimization decisions.

[0077] In step 113, a database containing economic environment parameters such as energy market prices and carbon tax unit prices is first established or accessed. This database needs to be updated regularly to ensure data timeliness and accuracy. Relevant numerical data can be obtained from pre-defined platforms, such as government public databases and industry trading platforms. These platforms typically provide data interfaces or APIs to facilitate automated data collection and integration. Numerical data obtained from pre-defined databases or platforms is integrated with sludge property data and process parameters to form a complete data set.

[0078] In step 114, relevant environmental protection policies and regulations are first collected and organized to form a pre-defined text information database. This text information may come from official government websites, industry associations, and other sources. The pre-defined text information is parsed to extract key information, such as policy terms, implementation dates, and impact areas. Natural language processing technology can be utilized during the parsing process to improve the accuracy and efficiency of information extraction. The parsed text data is then correlated with sludge attribute data, process parameters, and numerical data to form a multi-dimensional dataset. The correlation process must consider the logical and temporal relationships between the data.

[0079] In step 115, GIS technology is first used to geocode the sludge source area and record the geographic location of the sludge. The GIS system can provide map display and spatial analysis functions for the sludge source area. Key events and data in the sludge treatment process are timestamped to record time series information of the treatment stages. Timestamps can be used to analyze the temporal changes and patterns of the sludge treatment process. The geocoded information provided by the GIS is then integrated with the timestamped time series information to form spatiotemporal data. Spatiotemporal data can be used to analyze the geographical distribution characteristics of the sludge source area and the temporal changes of the treatment process, providing strong support for optimizing sludge resource utilization paths.

[0080] In an optional embodiment of the present invention, step 12 may include:

[0081] Step 121, normalizing the sludge attribute data, the process parameters, and the numerical data to obtain normalized data;

[0082] Step 122, performing vectorization processing on the text data to obtain a semantic vector;

[0083] Step 123, performing missing filling processing on the spatiotemporal data to obtain filled spatiotemporal data;

[0084] Step 124 , aligning the normalized data, the semantic vector, and the filled spatiotemporal data according to a preset dimension to obtain sludge sample data.

[0085] In this embodiment, step 121 is used to normalize the sludge attribute data (such as organic matter content, heavy metal concentration, moisture content, etc.), process parameters (such as reaction temperature, pH value, residence time, unit energy consumption, etc.) and numerical data in external dynamic data (such as energy market price, carbon tax unit price, etc.) to eliminate dimensional differences and enable data to be compared and analyzed on the same scale. Among them, the normalization method adopts the maximum and minimum standardization method to linearly map the original data to the [0,1] interval. For numerical data with missing values, linear interpolation is performed based on the adjacent time period data of the same sludge sample. Specifically, if the data of a certain time period is missing, linear interpolation is performed using the valid data of the adjacent time periods before and after it to estimate the missing value. The normalized data will have a unified scale, which is convenient for subsequent feature fusion and model training.

[0086] In step 122, the text data in the external dynamic data (such as the text of environmental protection policies and regulations) is vectorized to convert the text data into numerical data that can be processed by computers, that is, semantic vectors. The specific process includes: inputting the policy text into a pre-trained language model (such as BERT), and extracting the hidden layer output of the [CLS] tag as a global semantic vector. Through the attention mechanism of the BERT model, the keywords in the policy text (such as "banning sludge landfill", "prioritizing anaerobic digestion technology", etc.) are reinforced and encoded, thereby capturing the policy-oriented constraints on path selection. The obtained semantic vector can reflect the semantic information of the policy text and provide policy constraints for subsequent multi-agent reinforcement learning.

[0087] In step 123, the spatiotemporal data in the external dynamic data, such as the geographic coding of the sludge source area and the time series information of the treatment stage, are filled in to solve the problem of missing data and ensure the integrity and continuity of the data. The specific process includes: using a sliding time window mechanism for dynamic interpolation. Define the window size (such as w = 5, that is, 5 time periods before and after), and calculate the mean of the valid data in the window as the interpolation value. For example, if the pH value of timestamp t is missing, it is filled based on the arithmetic mean of the non-missing values ​​in the time period from t-5 to t+5. During the filling process, the correlation of spatiotemporal data needs to be retained to ensure that the filled data can reflect the spatiotemporal variation law of the sludge treatment process. The filled spatiotemporal data will have complete spatiotemporal information, providing a basis for the subsequent construction of multidimensional data tensors.

[0088] In step 124, the normalized sludge attribute data, process parameters, numerical external data, semantic vectors, and padded spatiotemporal data are aligned according to a preset dimensional structure. For example, the data for each sludge sample is organized into structured data containing a sample number, feature vector, timestamp, spatial encoding, and external factor vector. The aligned data is then integrated to form a complete sludge sample dataset. This dataset should contain all necessary information for subsequent multidimensional data tensor construction and model training. The resulting sludge sample data will have a unified format and structure, facilitating subsequent data processing and analysis.

[0089] In an optional embodiment proposed by the present invention, step 13 may include:

[0090] Step 131: classify the sludge sample data according to a preset dimension to obtain a plurality of classified data; wherein the preset dimension includes at least one of a sample dimension, a feature dimension, an external factor dimension, a time dimension, and a space dimension;

[0091] Step 132: Integrate the multiple classification data to obtain a multidimensional tensor, where the multidimensional tensor is represented as: T(i, j, k, l, m);

[0092] Where i, j, k, l, and m are natural numbers, and T(i, j, k, l, m) is the observed value of the jth feature of the i-th sample under the influence of the k-th external factor, in the l-th time period and the m-th region.

[0093] In step 131 of this embodiment, the sample dimension means that each sludge sample corresponds to a processing batch or record, and the sample dimension data identifies the source and processing batch of the sludge. For example, the sludge batches generated by a sewage treatment plant in a single day. The feature dimension includes sludge attribute data (such as organic matter content, heavy metal concentration, moisture content) and process parameters (such as reaction temperature, pH value, residence time, specific energy consumption). These characteristics reflect the physical and chemical properties of the sludge and the key parameters in the treatment process. The external factor dimension refers to the integration of dynamic external variables, including numerical data (such as energy market prices, carbon tax unit prices), text data (such as semantic vectors of environmental protection policy and regulations texts), etc. These external factors have an important impact on the choice of sludge resource utilization path. The time dimension refers to the division into multiple time slots according to the treatment stage, recording the continuous changes of various parameters during the sludge treatment process. The time dimension data helps to analyze the temporal evolution law of the sludge treatment process. The spatial dimension refers to the division based on the geographical code of the sludge source area, recording the geographical location information of the sludge. The spatial dimension data helps to analyze the geographical distribution characteristics of the sludge source area. Through the classification of the above five dimensions, the sludge sample data is split into multiple classification data sets, each of which corresponds to specific data under one dimension.

[0094] In step 132, a five-dimensional tensor is constructed based on the classified data set. Each dimension of the tensor corresponds to the five dimensions of sample, feature, external factor, time and space. Each element T(i, j, k, l, m) in the tensor represents an observation under the conditions of a specific sample (i), feature (j), external factor (k), time (l) and space (m). These observations may be sludge attribute values, process parameter values, external factor values ​​or a combination thereof. The classified data set is filled according to the structure of the tensor to form a complete multidimensional tensor. During the integration process, the accuracy and consistency of the data must be ensured to avoid data dislocation or omission.

[0095] Multidimensional tensors can simultaneously model the multidimensional interactive relationships of data in samples, time, space and external environment, revealing the intrinsic correlation between data; and multidimensional tensors also provide high-fidelity feature inputs for subsequent multi-agent reinforcement learning game optimization, which helps to generate more scientific and sustainable sludge resource utilization paths.

[0096] In an optional embodiment of the present invention, step 14 may include:

[0097] Step 141: Decompose the multidimensional tensor data to obtain a set of low-rank factor matrices;

[0098] Step 142: Dynamically update the time dimension factor matrix in the low-rank factor matrix set to obtain a decomposed factor matrix.

[0099] In step 141 of this embodiment, a tensor decomposition model with non-negative constraints and sparse regularization is used to transform the multidimensional tensor Decomposed into five groups of low-rank factor matrices A (1) 、A (2) 、A (3) , t (4) and A (5) , corresponding to samples, features, external factors, time and space dimensions respectively. Specifically, the decomposition objective function is defined as:

[0100]

[0101] in, represents the tensor reconstruction operator, ||·|| F is the Frobenius norm, and λ is the sparse regularization coefficient. It should be noted that the non-negative constraint (A (n) ≥0) ensures the physical interpretability of the factor matrix, such as the time dimension factor matrix A (4) Each column of can be regarded as the contribution weight of different time periods to the potential features; the sparse regularization term ||A (n) ||1 is used to remove redundant features and improve the generalization ability of the model.

[0102] In step 142, in order to adapt to the real-time changes in the external environment (such as policy changes, energy price fluctuations), an incremental learning strategy is used to calculate the time dimension factor matrix A. (4) Perform dynamic updates. For example, at fixed time intervals Δt (e.g., 1 hour), perform the following operations:

[0103] (1) Data window sliding: remove the data slice of the oldest time window (such as the first period) And add data slices for new time periods

[0104] (2) Local gradient descent update: fix other factor matrices A (1) ,A (2) ,A (3) ,A (5) , only for A (4) Perform iterative optimization and update the formula as follows:

[0105]

[0106] Where η is the learning rate, is the updated time window data. It can be understood that this method significantly reduces the computational overhead of full retraining while ensuring model stability.

[0107] It should be noted that the factor matrix obtained by decomposition contains the potential characteristics of the sludge resource utilization process. Specifically:

[0108] Sample dimension factor matrix A (1) : Each row represents the distribution of a single sample on each potential feature, which can be used to cluster similar processing batches.

[0109] Feature dimension factor matrix A (2) : Revealing the hidden correlation between sludge properties and process parameters, such as the synergistic effect between heavy metal concentration and reaction temperature.

[0110] External factor dimension factor matrix A (3) : Quantify the impact weight of dynamic variables such as policies and energy prices on the processing path. For example, the feature activation value corresponding to the keyword "carbon emission reduction" in the policy semantic vector is relatively high.

[0111] Time dimension factor matrix A (4) : Reflects the temporal evolution of potential characteristics in the treatment process, such as the increased preference for pyrolysis during the nighttime period with low electricity prices.

[0112] Spatial dimension factor matrix A (5) : Characterizes the impact of regional differences on the selection of treatment pathways, such as the lower weight of incineration processes in areas with high emission limits.

[0113] Through the above-mentioned tensor decomposition and dynamic update mechanism, this embodiment achieves efficient dimensionality reduction and feature extraction of multi-source heterogeneous data, providing high-fidelity feature input for subsequent multi-objective optimization.

[0114] In an optional embodiment proposed by the present invention, step 15 includes:

[0115] Step 151: input the decomposed factor matrix into a path optimization model for processing to obtain an intermediate result;

[0116] Step 152: Filter the intermediate results to obtain a resource optimization path set;

[0117] In step 151 of this embodiment, the path optimization model includes an economic agent, an environmental agent, and a social agent, which correspond to the goals of profit maximization, carbon emission minimization, and compliance maximization, respectively. It should be noted that each agent shares the same environmental state but makes independent decisions, and reaches a global optimal strategy combination through game interaction. In one possible implementation, the state space s of the economic agent is E include:

[0118] Real-time energy prices: normalized electricity or natural gas prices obtained from external dynamic data;

[0119] Sludge calorific value: calorific value per unit mass calculated based on sludge property data;

[0120] Tensor decomposition eigenvector f NTF : Step 14 decomposes the potential features related to economic benefits in the obtained feature matrix, such as process relevance weights.

[0121] Specifically, its action space It is defined as the resource recovery path selection, including three discrete actions: pyrolysis, composting, and incineration. For example, the reward function R E Designed to:

[0122] R E =E p -α·(E h +T c )

[0123] Among them, α is the cost penalty coefficient, which is determined by fitting historical data, R E is the reward value of the economic agent, E p is the expected profit, E h is the expected processing cost, T c is the transportation cost. It can be understood that this function drives the agent to choose a high-profit path by positively incentivizing profit growth and negatively suppressing cost expenditure.

[0124] Alternatively, the state space s of the environment agent Env include:

[0125] Real-time carbon emission intensity: carbon emissions per unit of processing volume calculated based on the type of processing technology and energy consumption;

[0126] Carbon tax unit price: normalized carbon tax price obtained from external dynamic data;

[0127] Historical average carbon emissions: the average carbon emissions within a sliding time window (such as the past 24 hours).

[0128] Its action space Defined as the carbon emission constraint threshold a Env ∈[0,1], is mapped to a specific carbon emission cap through the Sigmoid function. For example, the reward function R Env Designed to:

[0129] R Env =-β·(C e ×C t );

[0130] Among them, β is the environmental penalty coefficient, R Env is the reward value of the environment agent, C e is carbon emissions, C t It should be noted that the negative reward mechanism forces the agent to prioritize low-carbon emission processes, such as composting instead of incineration.

[0131] In one possible implementation, the state space s of the social agent is S include:

[0132] Policy semantic vector v policy : The policy text encoding generated by the BERT model in step 12;

[0133] Historical violation count: counts the number of times the recent processing path was penalized for violations.

[0134] Its action space Defined as the compliance score a S ∈[0,1], is converted into compliance probabilities of different paths through the Softmax function. For example, the reward function R S Designed to:

[0135] R S =γ·C s

[0136] Among them, γ is the compliance reward coefficient, R S is the reward value of the social agent, C s Compliance score. It can be understood that this function ensures that the decision complies with the policy constraints by amplifying the reward signal of the high-scoring path.

[0137] In step 152, each agent updates the action-value function based on the Q-learning algorithm and converges to a stable solution through the equilibrium strategy:

[0138] (1) Q-value function update:

[0139] Each agent selects action a according to the current state s i , get reward R after interaction i And transfer to the next state s ′ , the update formula is:

[0140]

[0141] Among them, Q i (s,a i ) indicates that agent i selects action a in state s i The value of R i is the immediate reward of agent i; η is the learning rate, δ is the discount factor; s ′ To perform action a i The next state to be transferred to; a i ′ Indicates that agent i is in state s ′ The actions that can be taken in Indicates that the agent is in the next state s′ The maximum expected cumulative reward under .

[0142] (2) Equilibrium strategy convergence:

[0143] Through iterative game, each agent strategy combination The equilibrium condition is met:

[0144]

[0145] in, Represent the action spaces of economic agent, environmental agent and social agent respectively; a E 、a Env 、a S Respectively represent any candidate actions of the economic agent, environmental agent, and social agent in their action spaces; They represent the optimal actions of economic agent, environmental agent and social agent under equilibrium strategy respectively.

[0146] For example, when the economic agent chooses the incineration path, the environmental agent may increase the carbon emission constraint threshold to offset its environmental costs, while the social agent dynamically adjusts the compliance score weight according to the policy vector.

[0147] Through the above multi-agent game mechanism, this embodiment realizes the dynamic trade-off and conflict resolution among multiple objectives, and obtains the resource optimization path set, which is recorded as The process includes:

[0148] Input graph structure (nodes represent locations, edges represent connection relationships) and starting and ending points;

[0149] Generate an initial path from the starting point, such as a single-node path, and store it in an open list;

[0150] Iteratively select paths from the open list (e.g., by lowest cost first), expand the adjacent nodes of their terminal nodes, and generate new path branches;

[0151] The new path cost is evaluated by the formula:

[0152] f(p)=g(p)+h(p)

[0153] Among them, g(p) is the actual cost from the starting point to the current node, such as the distance; h(p) is the heuristic function, such as the estimated value of the Euclidean distance to the end point.

[0154] If the expanded path does not reach the end point, it is re-added to the open list as a candidate path; if it has reached the end point, it is output as a feasible path.

[0155] Screening and termination: Invalid paths are removed through collision detection, such as grid map obstacle verification; the algorithm continues to iterate until the termination condition is met, such as finding N paths, the open list is empty, or resources are exhausted.

[0156] Furthermore, the present invention also provides steps for constructing the above framework, including:

[0157] (1) Framework initialization and parameter configuration

[0158] Alternatively, the construction of the multi-agent framework begins with the following initialization operations:

[0159] Economic agent: Initialize the Q value table dimension to |s E |×3 (state space size × number of actions), where the state space elements include real-time energy prices, sludge calorific value, and tensor decomposition feature vectors as mentioned above, and the actions are pyrolysis, composting, and incineration path selection.

[0160] Environmental agent: A neural network is used to approximate the continuous action space (carbon emission constraint threshold). The network input layer dimension matches the state space (carbon emission intensity, carbon tax unit price, historical carbon emission average), and the output layer is a single neuron representing the threshold.

[0161] Social Agent: Initializes the compliance score generator, maps the policy semantic vector and historical violation records into a 0-1 score value through a fully connected layer.

[0162] It should be noted that the learning rate η and the discount factor δ are set uniformly before training.

[0163] (2) Interaction Mechanism and Action Execution

[0164] In one possible implementation, the action execution process of each agent is as follows:

[0165] Action selection: The economic agent selects a path based on the ∈-greedy strategy (e.g., ∈=0.1 exploration probability); the environmental agent outputs the carbon emission threshold through a neural network; and the social agent generates a compliance score.

[0166] Environmental Simulator Feedback: Combine actions (a E ,a Env ,a S ) Input the simulator and calculate the multi-objective feedback:

[0167] Economic feedback: If path a E Carbon emissions exceed the threshold a env , then the profit will be deducted according to the excess proportion;

[0168] Compliance Verification: If score a S If it falls below a preset threshold, a violation is flagged and a penalty is triggered.

[0169] (3) Model updating and equilibrium convergence

[0170] Specifically, the update logic of each agent is designed differently as follows:

[0171] Economic Agent: Uses Tabular Q-learning to update the value of discrete actions. The update formula is as shown above, but is limited to path selection actions.

[0172] Environmental and social agents: Neural network parameters are updated using gradient descent, with the loss function being the mean squared error between the predicted action value and the target value.

[0173] It should be noted that the equilibrium convergence condition is verified through periodic strategy evaluation. When the strategy change rate of each agent is less than 1% in 10 consecutive iterations, it is judged to be converged.

[0174] In an optional embodiment of the present invention, step 16 may include:

[0175] Step 161: extract the resource optimization path set to obtain a first candidate solution and a second candidate solution;

[0176] Step 162: Determine the dominance relationship between the first candidate solution and the second candidate solution according to a preset dominance condition to determine a dominated solution set and a non-dominated solution set;

[0177] Step 163 : sort the non-dominated solution set according to the preset target priority to obtain the optimal utilization path data.

[0178] In step 161 of this embodiment, the candidate path set is recorded as Each solution Contains three objective function values, among which, is the profit of the kth path, which needs to be maximized and is calculated based on the difference between the processing benefit and cost;

[0179] is the carbon emission of the kth path, which needs to be minimized and is obtained by multiplying the process energy consumption and the carbon emission coefficient;

[0180] The compliance score of the kth path needs to be maximized and is dynamically generated by the social agent based on the policy semantic vector.

[0181] For example, if the profit of the pyrolysis path is 100,000 yuan, the carbon emissions are 500 kg, and the compliance score is 0.9, then its solution vector is p k =(10,500,0.9).

[0182] Extract the resource optimization path set to obtain any two solutions p iWith p j , namely the first candidate solution and the second candidate solution;

[0183] In step 162, for any two solutions p i With p j , if the following pre-set governing conditions are met:

[0184] (1) The economic objectives are no less than:

[0185] (2) Environmental objectives are no less than:

[0186] (3) The social objectives are not inferior to:

[0187] (4) At least one objective is strictly superior to: At least one of the three inequalities above is a strict inequality (> or <); then p i Dominate p j (denoted as ), and eliminate the dominated solution p j .

[0188] For example, if the solution p1 = (12,600, 0.8) and the solution p2 = (10,500, 0.9):

[0189] Economic target: 12 ≥ 10 (p1 is better);

[0190] Environmental target: 600 ≥ 500 (p2 is better);

[0191] Social goals: 0.8≤0.9 (p2 is better).

[0192] Since p1 and p2 do not dominate each other, both are retained.

[0193] As an alternative, the frontier solution set It is defined as the set of all candidate solutions that are not dominated by any other solution. The mathematical expression is:

[0194]

[0195] in, means "does not exist", that is, for the solution p k , if there is no other solution p m Dominate p k , then p k Belongs to the Pareto front. It can be understood that the solutions in this set cannot further optimize a certain objective without compromising other objectives, representing the optimal compromise surface of multi-objective optimization. Through the above process, the dominated solution set p is determined k and the non-dominated solution set p k ;

[0196] In step 163, the frontier solution set is sorted according to the preset target priority and then output. For example, if the priority is economy > environment > society, the preset target priority is:

[0197] (1) First priority: Press f econ Sort in descending order;

[0198] (2) Second priority: f econ Same solution, press f env Sort in ascending order;

[0199] (3) The third priority: f econ With f env The same solution, press f soc Sort in descending order.

[0200] It should be noted that this ranking mechanism allows decision makers to quickly locate the preferred solution based on actual needs (such as policy preferences or short-term benefit goals) without having to traverse the entire solution set.

[0201] Through the above process, this embodiment realizes the scientific screening and visual output of the multi-objective optimization solution set, and obtains the optimal utilization path data.

[0202] In an optional embodiment proposed by the present invention, the method for determining a sludge resource utilization path further includes:

[0203] Step 171, obtaining actual processing effect data;

[0204] Step 172: Input the actual processing effect data and the optimal utilization path data into a loss function for processing to obtain deviation data;

[0205] Step 173: updating the decomposed factor matrix according to the deviation data to obtain an updated factor matrix;

[0206] Step 174 : updating the path optimization model according to the updated factor matrix to obtain an updated path optimization model.

[0207] In this embodiment, the real-time feedback data update step is used to achieve dynamic closed-loop optimization of the model and strategy. Its core is to reverse-correct prediction deviations using actual processing performance data, ensuring the system's continuous adaptability to dynamic environments. Specifically, this step forms an iterative reinforcement loop of "data-model-decision" through the construction of a multi-objective loss function, gradient-driven parameter updates, and a strategy synchronization mechanism.

[0208] In step 171, the actual processing effect data includes the actual profit R real , actual carbon emissions R realand actual compliance score C real , corresponding to the implementation results of economic, environmental and social goals respectively. It should be noted that the data sources include:

[0209] Economic data: Obtain energy recovery revenue, processing costs, and transportation costs for the processing path from the financial system to calculate net profit;

[0210] Environmental data: Real-time monitoring of process energy consumption through IoT sensors, and calculation of total carbon emissions based on carbon emission coefficients;

[0211] Compliance data: compliance review results fed back by regulatory authorities, converted into a standardized score of 0-1.

[0212] In step 172, a weighted mean square error loss function is constructed to quantify the model prediction value (R pred , E pred , C pred ) Deviation from the actual value:

[0213]

[0214] Among them, R pred ,E pred ,C pred where α1, α2, and α3 represent the model's predicted profit, carbon emissions, and compliance score, respectively; α1, α2, and α3 represent the feedback weights for each objective. It should be noted that the weight coefficients are dynamically adjusted based on business priorities. For example, when environmental protection policies are tightened, α2 can be increased to strengthen carbon emission controls.

[0215] In step 173, the five factor matrices A of the dynamic tensor decomposition model are updated using the gradient descent method. (1) ,…,A (5) , the update formula is:

[0216]

[0217] Among them, η fb For feedback learning rate, the value range is limited to 0.001-0.1 to prevent gradient explosion. It can be understood that the update direction of the factor matrix is ​​determined by the partial derivative of the loss function with respect to each matrix, thereby minimizing the prediction deviation. For example, the time dimension factor matrix A (4) The update will correct the model's ability to predict the temporal evolution of process parameters.

[0218] In step 174, the updated factor matrix is ​​input into the path optimization model to trigger strategy retraining:

[0219] (1) Eigenvector reconstruction: based on the new factor matrix A (1) ,…,A (5) Recalculate the tensor eigenvector fNTF , replace the original state input of the economic agent;

[0220] (2) Q-value table reset: retain 80% of the historical Q-values ​​as prior knowledge, and reinitialize the remaining 20% ​​based on the new feature distribution to balance experience inheritance and exploration capabilities;

[0221] (3) Strategy weight adjustment: The carbon emission constraint threshold generation network of the environmental agent and the compliance scoring network parameters of the social agent are updated synchronously to adapt to the latest tensor features.

[0222] It should be noted that this synchronization mechanism avoids strategy shock caused by directly resetting the agent and ensures decision consistency.

[0223] Through the above process, the path optimization model is updated.

[0224] like Figure 2 As shown, the embodiment of the present invention further provides a sludge resource utilization path determination device 20, comprising:

[0225] An acquisition module 21 is used to acquire raw data, wherein the raw data includes at least one of sludge attribute data, process parameters and external dynamic data;

[0226] The processing module 22 is configured to pre-process the raw data to obtain sludge sample data; classify and integrate the sludge sample data to obtain a multidimensional tensor; decompose and dynamically update the multidimensional tensor to obtain a decomposed factor matrix; and input the decomposed factor matrix into a path optimization model for processing to obtain a resource optimization path set.

[0227] The determination module 23 is configured to perform optimal verification processing on the resource optimization path set to obtain optimal utilization path data.

[0228] Optionally, the acquisition module 21 is specifically configured to:

[0229] Acquiring the sludge property data through a sensor array or laboratory testing, wherein the sludge property data includes at least one of organic matter content, heavy metal concentration, and moisture content;

[0230] Obtaining the process parameters through real-time recording of the sludge treatment equipment monitoring system, wherein the process parameters include at least one of reaction temperature, pH value, residence time, and unit energy consumption;

[0231] Obtaining numerical data from the external dynamic data through a preset database or a preset platform;

[0232] Obtaining text data from the external dynamic data through preset text information;

[0233] The spatiotemporal data in the external dynamic data is obtained by associating the geographic information system with the timestamp tag.

[0234] Optionally, the processing module 22 is specifically configured to:

[0235] Normalizing the sludge attribute data, the process parameters, and the numerical data to obtain normalized data;

[0236] Performing vectorization processing on the text data to obtain a semantic vector;

[0237] Performing missing filling processing on the spatiotemporal data to obtain filled spatiotemporal data;

[0238] The normalized data, the semantic vector and the filled spatiotemporal data are aligned according to a preset dimension to obtain sludge sample data.

[0239] Optionally, the processing module 22 is further specifically configured to:

[0240] Classifying the sludge sample data according to preset dimensions to obtain a plurality of classified data; wherein the preset dimensions include at least one of a sample dimension, a feature dimension, an external factor dimension, a time dimension, and a space dimension;

[0241] Integrating the multiple classification data to obtain a multidimensional tensor, wherein the multidimensional tensor is represented as: T(i, j, k, l, m);

[0242] Where i, j, k, l, and m are natural numbers, and T(i, j, k, l, m) is the observed value of the jth feature of the i-th sample under the influence of the k-th external factor, in the l-th time period and the m-th region.

[0243] Optionally, the processing module 22 is further specifically configured to:

[0244] Decomposing the multidimensional tensor data to obtain a set of low-rank factor matrices;

[0245] The time dimension factor matrix in the low-rank factor matrix set is dynamically updated to obtain a decomposed factor matrix.

[0246] Optionally, the path optimization model includes at least one of an economic agent, an environmental agent, and a social agent;

[0247] The state space of the economic agent includes at least one of real-time energy price, sludge calorific value and tensor decomposition eigenvector; the action space of the economic agent includes at least one of pyrolysis action, composting action and incineration action; the reward function of the economic agent is: R E=E p -α·(E h +T c ), α is the cost penalty coefficient, R E is the reward value of the economic agent, E p is the expected profit, E h is the expected processing cost, T c For transportation costs;

[0248] The state space of the environmental agent includes at least one of real-time carbon emission intensity, carbon tax unit price and historical carbon emission average; the action space of the environmental agent includes carbon emission constraint threshold; the reward function of the environmental agent is: R Env =-β·(C e ×C t ), β is the environmental penalty coefficient, R Env is the reward value of the environment agent, C e is carbon emissions, C t is the unit price of carbon tax;

[0249] The state space of the social agent includes at least one of a semantic vector and a number of historical violations; the action space of the social agent includes a compliance score; and the reward function of the social agent is: R S =γ·Compliance score, γ is the compliance reward coefficient, R S is the reward value of the social agent, C s Score compliance.

[0250] Optionally, the determining module 23 is specifically configured to:

[0251] Extracting the resource optimization path set to obtain a first candidate solution and a second candidate solution;

[0252] Determine the dominance relationship between the first candidate solution and the second candidate solution according to a preset dominance condition to determine a dominated solution set and a non-dominated solution set;

[0253] The non-dominated solution set is sorted according to the preset target priority to obtain the optimal utilization path data.

[0254] Optionally, the sludge resource utilization path determination device 20 further includes:

[0255] The updating module 24 is used to obtain actual processing effect data; input the actual processing effect data and the optimal utilization path data into the loss function for processing to obtain deviation data; based on the deviation data, update the decomposed factor matrix to obtain an updated factor matrix; based on the updated factor matrix, update the path optimization model to obtain an updated path optimization model.

[0256] It should be noted that this device is a device corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0257] like Figure 3 As shown, an embodiment of the present invention further provides a computing device 30, including a processor 31, a memory 32, and a program or instruction stored in the memory 32 and executable on the processor 31. When the program or instruction is executed by the processor 31, the various processes of the above-mentioned embodiment of the method for determining a sludge resource utilization path are implemented, and the same technical effects are achieved. To avoid repetition, they are not described here. It should be noted that the computing device in the embodiment of the present invention includes the above-mentioned mobile electronic device and non-mobile electronic device.

[0258] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0259] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0260] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0261] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0262] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0263] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard disk, ROM, RAM, a magnetic disk, or an optical disk.

[0264] In addition, it should be noted that, in the apparatus and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. Moreover, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it will be understood that all or any steps or components of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or a network of computing devices in hardware, firmware, software or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0265] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general-purpose device. Therefore, the purpose of the present invention can also be achieved simply by providing a program product containing program code for implementing the method or device. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but do not necessarily need to be performed in chronological order. Certain steps can be performed in parallel or independently of each other.

[0266] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for determining a sludge resource utilization path, characterized in that: include: Acquiring raw data, wherein the raw data includes at least one of sludge attribute data, process parameters, and external dynamic data; Preprocessing the raw data to obtain sludge sample data; Classifying and integrating the sludge sample data to obtain a multidimensional tensor; Decomposing and dynamically updating the multidimensional tensor to obtain a decomposed factor matrix; Inputting the decomposed factor matrix into a path optimization model for processing to obtain a resource optimization path set; An optimal verification process is performed on the resource optimization path set to obtain optimal utilization path data.

2. The method for determining a sludge resource utilization path according to claim 1, characterized in that: The obtaining of original data includes: Acquiring the sludge property data through a sensor array or laboratory testing, wherein the sludge property data includes at least one of organic matter content, heavy metal concentration, and moisture content; Obtaining the process parameters through real-time recording of the sludge treatment equipment monitoring system, wherein the process parameters include at least one of reaction temperature, pH value, residence time, and unit energy consumption; Obtaining numerical data from the external dynamic data through a preset database or a preset platform; Obtaining text data from the external dynamic data through preset text information; The spatiotemporal data in the external dynamic data is obtained by associating the geographic information system with the timestamp tag.

3. The method for determining a sludge resource utilization path according to claim 2, characterized in that: The raw data is preprocessed to obtain sludge sample data, including: Normalizing the sludge attribute data, the process parameters, and the numerical data to obtain normalized data; Performing vectorization processing on the text data to obtain a semantic vector; Performing missing filling processing on the spatiotemporal data to obtain filled spatiotemporal data; The normalized data, the semantic vector and the filled spatiotemporal data are aligned according to a preset dimension to obtain sludge sample data.

4. The method for determining a sludge resource utilization path according to claim 1, characterized in that: The sludge sample data is classified and integrated to obtain a multidimensional tensor, including: Classifying the sludge sample data according to preset dimensions to obtain a plurality of classified data; wherein the preset dimensions include at least one of a sample dimension, a feature dimension, an external factor dimension, a time dimension, and a space dimension; Integrating the multiple classification data to obtain a multidimensional tensor, wherein the multidimensional tensor is represented as: T(i, j, k, l, m); Where i, j, k, l, and m are natural numbers, and T(i, j, k, l, m) is the observed value of the jth feature of the i-th sample under the influence of the k-th external factor, in the l-th time period and the m-th region.

5. The method for determining a sludge resource utilization path according to claim 1, characterized in that: The multidimensional tensor is decomposed and dynamically updated to obtain a decomposed factor matrix, including: Decomposing the multidimensional tensor data to obtain a set of low-rank factor matrices; The time dimension factor matrix in the low-rank factor matrix set is dynamically updated to obtain a decomposed factor matrix.

6. The method for determining a sludge resource utilization path according to claim 1, characterized in that: The path optimization model includes at least one of an economic agent, an environmental agent, and a social agent; The state space of the economic agent includes at least one of real-time energy price, sludge calorific value and tensor decomposition eigenvector; the action space of the economic agent includes at least one of pyrolysis action, composting action and incineration action; the reward function of the economic agent is: R E =E p -α·(E h +T c ), α is the cost penalty coefficient, R E is the reward value of the economic agent, E p is the expected profit, E h is the expected processing cost, T c For transportation costs; The state space of the environmental agent includes at least one of real-time carbon emission intensity, carbon tax unit price and historical carbon emission average; the action space of the environmental agent includes carbon emission constraint threshold; the reward function of the environmental agent is: R Env =-β·(C e ×C t ), β is the environmental penalty coefficient, R Env is the reward value of the environment agent, C e is carbon emissions, C t is the unit price of carbon tax; The state space of the social agent includes at least one of a semantic vector and a number of historical violations; the action space of the social agent includes a compliance score; and the reward function of the social agent is: R S =γ·C s , γ is the compliance reward coefficient, R S is the reward value of the social agent, C s Score compliance.

7. The method for determining a sludge resource utilization path according to claim 1, characterized in that: Performing optimal verification processing on the resource optimization path set to obtain optimal utilization path data includes: Extracting the resource optimization path set to obtain a first candidate solution and a second candidate solution; Determine the dominance relationship between the first candidate solution and the second candidate solution according to a preset dominance condition to determine a dominated solution set and a non-dominated solution set; The non-dominated solution set is sorted according to the preset target priority to obtain the optimal utilization path data.

8. The method for determining a sludge resource utilization path according to claim 1, characterized in that: Also includes: Obtain actual processing effect data; Inputting the actual processing effect data and the optimal utilization path data into a loss function for processing to obtain deviation data; performing an updating process on the decomposed factor matrix according to the deviation data to obtain an updated factor matrix; The path optimization model is updated according to the updated factor matrix to obtain an updated path optimization model.

9. A device for determining a sludge resource utilization path, characterized in that: include: An acquisition module, configured to acquire raw data, wherein the raw data includes at least one of sludge attribute data, process parameters, and external dynamic data; A processing module, used for preprocessing the raw data to obtain sludge sample data; Classifying and integrating the sludge sample data to obtain a multidimensional tensor; Decomposing and dynamically updating the multidimensional tensor to obtain a decomposed factor matrix; inputting the decomposed factor matrix into a path optimization model for processing to obtain a resource optimization path set; The determination module is used to perform optimal verification processing on the resource optimization path set to obtain optimal utilization path data.

10. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method according to any one of claims 1 to 8 is performed.

Images