Highway solid waste treatment equipment energy consumption monitoring and resource utilization optimization system

Abstract

The application relates to the technical field of solid waste treatment equipment energy consumption monitoring and optimization, and discloses a highway solid waste treatment equipment energy consumption monitoring and resource utilization optimization system, which comprises the following: a data fusion module acquires fused multi-source data and generates a probabilistic composition vector of materials; a market insight module quantifies liquidity based on market data and generates the most valuable target product combination; a reverse planning module reversely searches a process with the target product as the terminal point, generates a treatment chain and evaluates a success probability; a strategy decision module calculates the expected strategy profit of each treatment chain and determines a globally optimal execution strategy; and a scheduling execution center resolves the optimal strategy into a device operation instruction set and issues the same to an industrial control system. The application finally solves the optimal execution strategy with the maximum expected strategy profit by calculating the expected achievable income containing the success probability, market liquidity discount and total execution cost, so that each production can target the optimal economic return.

Description

Technical Field

[0001] This invention relates to the field of energy consumption monitoring and optimization technology for solid waste treatment equipment, specifically a system for energy consumption monitoring and resource utilization optimization of highway solid waste treatment equipment. Background Technology

[0002] Highway solid waste, as a major type of construction waste, has a complex composition and significant batch-to-batch variations. This type of waste stream contains metals of varying grades, recycled aggregates of different specifications, and other usable materials. Therefore, efficient resource recovery of this type of solid waste is not only a necessary requirement for environmental protection but also holds considerable economic potential.

[0003] Currently, the mainstream application solution in the industry is the traditional processing production line based on physical separation technology. These production lines are usually composed of a series of equipment such as crushers, screening machines, magnetic separators, and air separators connected in a fixed sequence. After the solid waste raw materials to be processed are fed into the production line, they are processed along this preset and unchanging process path. The main responsibility of the operators is to monitor the operating status of the equipment and ensure that the materials can pass through this fixed process smoothly.

[0004] However, existing energy consumption monitoring and utilization optimization systems often struggle to quantify the market monetization potential of specific products due to fluctuations in the real-time prices of non-renewable resources, especially when processing materials and producing products. The systems must handle all incoming materials, leading to unnecessary energy and equipment losses when processing low-value materials. Furthermore, they lack time-based optimization capabilities, failing to proactively utilize periods of low electricity prices and failing to convert idle time while waiting for market opportunities into effective production preparation activities, resulting in losses in both economic benefits and operational efficiency. Therefore, this invention provides an energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment. This system solves the problems of static solid waste treatment strategies in existing technologies, which are difficult to adapt to dynamic market changes and lack the ability to quantify uncertainties in composition, resulting in poor economic benefits.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment, comprising:

[0007] The data fusion module is used to acquire and fuse multi-source heterogeneous data in real time. The multi-source heterogeneous data includes solid waste treatment equipment status data, material flow data of batches of solid waste to be treated, and external market data. The module also parses the material flow data to generate a probabilistic component vector.

[0008] The market insight module is used to quantitatively calculate the market liquidity index of each potential productive renewable resource based on the external market data provided by the data fusion module, and dynamically screen and generate target product combinations under the current market environment.

[0009] The reverse planning module is used to perform reverse path search in a preset process knowledge graph library with the target product combination as the planning target, generate at least one dynamic processing chain for the target product combination, and evaluate the success probability of executing each dynamic processing chain to process a specified batch of solid waste in combination with the probabilistic component vector.

[0010] The strategy decision module is used to calculate the expected strategy profit for each generated dynamic processing chain and its corresponding success probability, and to determine the optimal execution strategy with the maximum expected strategy profit.

[0011] The scheduling and execution center is used to parse the optimal execution strategy into a structured set of operation instructions containing equipment control parameters and execution timing, and then send it to the industrial control system.

[0012] Preferably, the data fusion module includes:

[0013] The data acquisition unit is used to acquire the status data of the solid waste treatment equipment, the material flow data of the batch of solid waste to be treated, and the external market data through industrial standard protocols, online rapid component analysis devices, and application programming interfaces, respectively.

[0014] The probabilistic component vector generation unit is used to input the material flow data of the batch of solid waste to be processed acquired by the data acquisition unit into a pre-trained material composition inference model to output the probability distribution of the content of each recyclable resource, thereby generating a probabilistic component vector.

[0015] The data preprocessing and synchronization unit is used to clean and normalize the solid waste treatment equipment status data, the material flow data of the batch of solid waste to be treated, and the external market data acquired by the data acquisition unit, and to assign a unified timestamp to align them in the time dimension, forming a synchronized time-series data stream.

[0016] Preferably, the market insight module includes:

[0017] The market liquidity quantification unit is used to calculate the market liquidity index for each potentially productive renewable resource based on the trading volume, trading frequency and bid-ask spread in the external market data.

[0018] The value liquidity weighted scoring unit is used to combine the market price of the renewable resources and the market liquidity index to calculate the value liquidity score of the renewable resources through a weighted summation model.

[0019] The target product combination generation unit is used to determine and generate a target product combination from candidate resources based on the value mobility score and according to a preset screening logic.

[0020] Preferably, the formula for calculating the value liquidity score is:

[0021] ;

[0022] in, It is a resource At any moment The overall score; It is a resource At any moment Market price after normalization; It is a resource At any moment Market liquidity indicators after normalization; and These are preset weighting coefficients.

[0023] Preferably, the reverse planning module includes:

[0024] The reverse path search unit is used to recursively trace back to the initial process node with raw solid waste as input from the final process operation node that can directly produce the target product in the process knowledge graph database, thereby generating a dynamic processing chain.

[0025] The success probability assessment unit is used to calculate the success probability by comparing the minimum material requirements of the first process operation of the dynamic processing chain with the material characteristics of the batch of solid waste to be treated as described by the probabilistic component vector.

[0026] Preferably, the success probability assessment unit is specifically used for:

[0027] From the probabilistic component vector, a probability distribution vector representing the availability of various related resources in the batch of solid waste to be processed is inferred. Each element in the probability distribution vector is a probability distribution used to describe the uncertainty of the availability of the corresponding resource.

[0028] Extract a deterministic threshold vector from the dynamic processing chain to determine the minimum required quantity for the first process operation. Each element in the deterministic threshold vector is a definite value.

[0029] The success probability is determined by calculating the probability that the available resource quantity described by the probability distribution vector is greater than or equal to the minimum required quantity defined by the deterministic threshold vector.

[0030] Preferably, the strategy decision module includes:

[0031] The expected achievable revenue calculation unit is used to integrate the success probability, the market value of the product, and market liquidity to estimate the expected achievable revenue for each potential execution strategy.

[0032] The total execution cost calculation unit is used to comprehensively calculate the energy cost, operating cost, and warehousing and opportunity cost generated during the execution of each potential execution strategy in order to determine the total execution cost.

[0033] A global optimal strategy solver is used to calculate the expected strategy profit by subtracting the total execution cost from the expected achievable profit, and to traverse all potential execution strategies to determine the optimal execution strategy with the maximum expected strategy profit.

[0034] Preferably, the scheduling execution center includes:

[0035] The automatic job instruction set generation unit is used to decompose the dynamic processing chain in the optimal execution strategy into process operation sequences, and query the equipment identifier and key operating parameters for each process operation sequence according to the process knowledge graph library to generate a structured job instruction set.

[0036] The interface unit of the field control system is used to send the structured set of operation instructions to the industrial control system when the planned start time of the optimal execution strategy arrives, and to receive execution status feedback for real-time monitoring.

[0037] The forward-looking capacity management unit is used to proactively generate and issue preparatory or maintenance instructions related to the dynamic processing chain during the waiting period from the current time to the planned start time when the planned start time of the optimal execution strategy is later than the current time.

[0038] Preferably, the system further includes an operation visualization module, which includes:

[0039] The decision-making panoramic data display unit is used to graphically display the external market data, the solid waste treatment equipment status data, and the optimal execution strategy information in real time on the monitoring dashboard.

[0040] The decision process traceability unit is used to respond to a query for a selected strategy, retrieve and display the calculation basis of the strategy during the decision-making process, the calculation basis including a sorted list of expected strategy profit scores of candidate strategies, and the component values ​​of the expected achievable revenue and total execution cost of the selected strategy.

[0041] The manual intervention and strategy configuration unit is used to present the optimal execution strategy as pending review so that operations and management personnel can perform approval, rejection or manual selection operations.

[0042] It also provides a method for monitoring energy consumption and optimizing resource utilization of highway solid waste treatment equipment, including the following steps:

[0043] Real-time acquisition and fusion of multi-source heterogeneous data, including status data of multi-source heterogeneous solid waste treatment equipment, material flow data of batches of solid waste to be treated, and external market data, and parsing the material flow data to generate probabilistic component vectors;

[0044] Based on the external market data, the market liquidity index of each potential productive renewable resource is quantitatively calculated, and the target product combination is dynamically screened and generated under the current market environment.

[0045] Using the target product combination as the planning target, a reverse path search is performed in the preset process knowledge graph library to generate at least one dynamic processing chain for the target product combination, and the success probability of executing each dynamic processing chain to process a specified batch of solid waste is evaluated by combining the probabilistic component vector.

[0046] For each generated dynamic processing chain and its corresponding success probability, calculate the expected strategy profit, and determine the optimal execution strategy with the maximum expected strategy profit.

[0047] The optimal execution strategy is parsed into a structured set of operation instructions containing equipment control parameters and execution timing, and then sent to the industrial control system.

[0048] This invention provides an energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment. It has the following beneficial effects:

[0049] 1. This invention, by setting up a market insight module, not only considers the market price of recycled resources but also introduces market liquidity indicators for weighted scoring, screening out target product combinations that combine high value and high liquidity. The strategy decision module calculates the expected achievable revenue, including success probability and market liquidity discount, as well as the total execution cost, including dynamic electricity prices and opportunity costs, ultimately solving for the optimal execution strategy with the maximum expected strategy profit. This ensures that every production activity aims at the globally optimal economic return, thereby transforming solid waste treatment from a cost center into a profit center.

[0050] 2. This invention uses probabilistic component vectors to mathematically model the uncertainty of materials, enabling the system to quantitatively assess process risks during the planning stage. Furthermore, through the reverse planning module, a dynamic processing chain is generated based on the dynamically changing combination of target products, allowing the processing technology to flexibly match market demand and material conditions.

[0051] 3. This invention guides the system to perform high-energy-consuming processes during periods of low electricity prices by integrating the calculation of total execution cost with real-time time-of-use electricity prices in the strategy decision-making module. This directly reduces energy costs in the production process. The forward-looking capacity management unit set up in the scheduling execution center can effectively utilize the waiting period generated by market timing decisions to arrange pre-maintenance of equipment, pre-transfer of materials, and other preparatory work, turning idle time into effective production preparation time, shortening process connection delays, and improving the comprehensive utilization rate of equipment and overall production efficiency. Attached Figure Description

[0052] Figure 1 This is a system architecture diagram of the present invention;

[0053] Figure 2 This is a flowchart of the method steps of the present invention;

[0054] Figure 3 This is a schematic diagram of the reverse path search process of the present invention;

[0055] Figure 4 This is a schematic diagram of the expected strategy profit calculation model of the present invention. Detailed Implementation

[0056] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] See attached document Figure 1 , Figure 1 This is a schematic diagram of a system architecture according to an embodiment of the present invention. The present invention provides an energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment, comprising:

[0058] The data fusion module is used to acquire and integrate data required for decision-making from multiple heterogeneous data sources. The data fusion module is configured with communication interfaces with the industrial control network at the solid waste treatment site, online analysis devices, and external market data service providers. It is used to receive and parse internal equipment status data, material flow data, and external market data, and to perform preprocessing and time-series alignment operations on the acquired data to form a standardized data stream for other modules in the system to call.

[0059] The Market Insights module receives standardized market data from the Data Fusion module. This module has a built-in value liquidity weighted scoring model to quantitatively assess the commercial attractiveness of various renewable resources. It calculates and generates a Target Product Portfolio (TPP), which defines the production target with the highest economic benefits under current market conditions, and outputs it to the Reverse Planning module.

[0060] The reverse planning module takes the Target Product Combination (TPP) output by the Market Insight module as input. Based on its internally built process knowledge graph, this module searches for reverse paths with the target product as the endpoint, dynamically generating one or more technically feasible Dynamic Processing Chains (DPCs). Simultaneously, this reverse planning module combines material flow data to quantitatively evaluate the success probability of each DPC, and transmits the generated DPCs and their corresponding success probabilities to the strategy decision module.

[0061] The strategy decision-making module is used to select the best execution strategy from all candidate strategies. This module receives the Dynamic Processing Chain (DPC) and its success probability generated by the inverse programming module. Combined with market data, it uses a built-in Expected Strategy Profit (ESP) model to quantify the economic benefits of each potential strategy (a combination of processing chain and execution time). Its final output is the optimal strategy with the maximum expected strategy profit, providing a deterministic decision-making basis for scheduling execution.

[0062] The scheduling and execution center receives the optimal strategy output from the strategy decision module. The scheduling and execution center transforms the strategy decision into a set of specific and executable work instructions. These work instructions are sent to the central control system or manufacturing execution system (MES) at the solid waste treatment site through an interface to drive the operation of physical equipment. The scheduling and execution center also includes a forward-looking capacity management unit for performing preparatory tasks related to the future optimal strategy.

[0063] The operations visualization module is used to display equipment status, market conditions, the decision-making process and final results of the current optimal strategy to managers in real time. This operations visualization module provides necessary manual intervention interfaces, allowing the strategy generated by the system to be reviewed or adjusted under specific circumstances.

[0064] The data fusion module is used to acquire, parse, and integrate all the data required for decision-making from multiple heterogeneous data sources in real time, providing unified, clean, and time-aligned data support for subsequent analysis and decision-making modules. This data fusion module includes a data acquisition unit, a probabilistic component vector generation unit, and a data preprocessing and synchronization unit.

[0065] The data acquisition unit is responsible for establishing and maintaining stable communication connections with data sources both inside and outside the system. For internal equipment status data, this unit communicates directly with the programmable logic controllers (PLCs) or supervisory control systems (SCADA) in the solid waste treatment plant via industry-standard protocols such as OPC-UA (Open Platform Unified Communication Architecture) or Modbus TCP / IP to obtain real-time operating power, load rate, processing efficiency, and other status parameters of each treatment device. For material flow data, the unit receives raw data from online rapid component analysis devices on the feed conveyor belt. These devices can be X-ray fluorescence (XRF) spectrometers, near-infrared (NIR) spectrometers, or hyperspectral imaging systems. For external market data, the unit periodically obtains time-of-use electricity prices, spot and futures prices of various renewable resources, and related market micro-liquidity data from designated energy trading data service providers and commodity information platforms by calling RESTful API interfaces.

[0066] The probabilistic composition vector generation unit transforms the raw, high-dimensional sensor data output from the online composition analysis device into a structured data format capable of quantifying uncertainties. Due to the non-uniformity and complexity of highway solid waste composition, deterministic composition analysis results are difficult to obtain in real time and are costly. By employing a probabilistic description method, this probabilistic composition vector generation unit receives the raw analysis data and inputs it into a pre-trained material composition inference model. The model outputs the probability distribution of the content or grade of each potential valuable resource. For each batch B of solid waste to be processed, the probabilistic composition vector generation unit generates a probabilistic composition vector, mathematically expressed as:

[0067] ;

[0068] in, This represents the probabilistic component vector of solid waste batch B; Indicates the first Types of recyclable resources, such as copper, aluminum, and recycled aggregates of specific particle sizes; Representation and Resources The corresponding probability distribution is used to describe the uncertainty of the quality fraction or grade of the resource in batch B. In one specific embodiment, It can be a normal distribution , where parameters The parameter represents the mathematical expectation (most likely value) of this resource content. Its variance represents the uncertainty or range of fluctuation of the content value.

[0069] The data preprocessing and synchronization unit is responsible for cleaning, standardizing, and aligning the collected multi-source heterogeneous raw data to form a time-series dataset that can be directly used by upper-layer applications. This unit first cleans the data, including identifying and processing outliers or outliers in the data stream, and using methods such as linear interpolation or forward / backward imputation to handle missing data points. Subsequently, this unit performs normalization processing on data of different dimensions, such as using min-max normalization to map all data to intervals to eliminate the influence of different physical dimensions on subsequent model calculations. This unit assigns a uniform high-precision timestamp to all incoming data points to ensure that all data from equipment, materials, and markets are strictly aligned in the time dimension, forming a synchronized, multivariate time-series data stream, which serves as the final output of the data fusion module.

[0070] The market insight module receives standardized market data from the data fusion module and identifies the most economically valuable and commercially viable production targets. This module includes a market liquidity quantification unit, a value liquidity weighted scoring unit, and a target product portfolio generation unit.

[0071] The market liquidity quantification unit is used to transform market liquidity into specific, calculable metrics. This unit receives micro-market data from the data fusion module, specifically including real-time trading volume, trading frequency, bid-ask spread (or bid-ask spread) of a particular renewable resource on major trading platforms, and order book depth. The unit aggregates this multi-dimensional raw data into a single market liquidity indicator using a pre-defined synthesis function. One specific implementation method involves a market liquidity indicator that is positively correlated with trading volume and frequency, and negatively correlated with bid-ask spread. Its mathematical expression can be:

[0072] ;

[0073] in, Representative Resources At any moment Market liquidity indicators; It is a comprehensive weighted function; It is a resource At any moment Cumulative trading volume within; It is a resource At any moment Transaction frequency within; It is a resource The real-time buy and sell price spread.

[0074] The value liquidity-weighted scoring unit is used to comprehensively score each potentially productive renewable resource. This unit receives the real-time market price of the resource and a liquidity index calculated by the market liquidity quantification unit. To eliminate the influence of different index dimensions, the value liquidity-weighted scoring unit first normalizes the input price and liquidity index, for example, using a min-max normalization method to map them to intervals. Then, it calculates the final value liquidity score for each resource using a weighted summation model. :

[0075] ;

[0076] in, It is a resource At any moment The overall score; It is a resource At any moment Market price after normalization; It is a resource At any moment Market liquidity indicators after normalization; and These are preset weighting coefficients that satisfy... These represent the relative importance of price and liquidity in the final score, respectively. These two weighting coefficients can be adjusted by the system administrator based on market strategies.

[0077] The target product combination generation unit receives all candidate resources and their corresponding value liquidity scores. The system then determines the optimal set of production targets based on a preset filtering logic. In one implementation, a threshold method is used to collect all targets with scores higher than a preset threshold. In another implementation, the target product combination generation unit uses a ranking method to select the top-scoring resources. The resources are used to generate a dynamically updated TargetProduct Portfolio, which is mathematically expressed as follows:

[0078] ;

[0079] in, Represents the moment Dynamically generated combinations of target products; Represents a time variable; represents the first... Types of recyclable resources, such as copper, aluminum, and recycled aggregates of specific particle sizes; Representing resources At any moment Value-Liquidity Score This represents the preset value-liquidity score threshold.

[0080] The target product combination It includes one or more sets of renewable resources that the system deems most worthwhile to produce under the current market conditions. This dynamically generated combination will be output as an instruction.

[0081] See attached document Figure 3 , Figure 3 This is a schematic diagram of a reverse planning module according to an embodiment of the present invention. The reverse planning module receives a target product combination (TPP) determined by a market insight module. The reverse planning module includes a process knowledge graph library, a reverse path search unit, and a success probability assessment unit.

[0082] The process knowledge graph is the foundation for path planning in this reverse planning module. It's a professional knowledge model built in the form of a graph database. In this graph, each "node" represents a specific process operation (e.g., 50mm bar screen crushing, high-intensity magnetic separation, eddy current separation, etc.). Each node stores detailed attributes of that process operation, including the required input material characteristics (such as particle size range and composition requirements), the characteristics of the output material, and related energy consumption models and processing efficiencies. The "edges" in the graph represent the connections between processes, indicating that the output material of one node can be used as the input of another node. This knowledge graph comprehensively digitizes the capabilities of all equipment and the possibilities of process combinations within the solid waste treatment plant.

[0083] After receiving the Target Product Combination (TPP), the reverse path search unit searches for each target product in the combination. A path search is performed in the process knowledge graph. This path search process is in reverse: it first locates paths that can directly produce the target product. The final process operation node is then identified. Using the input material characteristics required by this node as a new target, the process continues recursively to find the preceding process operation node that can produce this intermediate material. This process continues until the initial process node with the raw solid waste as input is reached. This search process ultimately generates one or more complete process operation sequences from the raw solid waste to the target product. This complete process operation sequence is defined as a Dynamic Processing Chain, and its formal expression is as follows:

[0084] ;

[0085] in, Indicates the first A dynamic processing chain; This represents the total number of process operations that make up this specific dynamic processing chain DPCⱼ; This indicates the specific process operations that include equipment selection and parameter settings; This indicates the components of the dynamic processing chain. A series of sequentially executed process operations.

[0086] The success probability assessment unit is used to quantify the inherent risks of executing the Dynamic Processing Chain (DPC). This is due to the inherent uncertainty of solid waste components (the probabilistic component vector generated by the data fusion module). As described, DPC does not guarantee success when dealing with a specific batch of materials. This success probability assessment unit compares the initial process requirements of the DPC with the material characteristics of the batch B to be processed and calculates its success probability. Specifically, the prerequisite for the success of any DPC is that the batch B to be processed must meet its first process operation. The minimum material requirements. This success probability assessment unit calculates the success probability using the following formula. :

[0087] ;

[0088] in, This indicates that a dynamic processing chain is executed when processing batch B. The probability of success; It is a probability calculation function; It is a vector representing the probabilistic component vector. The available quantity of various related resources in batch B inferred from the vector, where each element is a probability distribution; It is a vector representing the execution of the first process operation. The minimum required quantities of various resources are a deterministic threshold vector. The calculation of this success probability involves... The joint probability density function of each resource quantity satisfies Perform multiple integrations over the given region.

[0089] See attached document Figure 4 , Figure 4 This is a schematic diagram of an expected strategy profit calculation model according to an embodiment of the present invention. The strategy decision module performs a comprehensive and quantitative economic benefit evaluation of all potential execution strategies and selects the only strategy that can achieve the globally optimal economic return. The strategy decision module receives multiple sets of dynamic processing chains (DPCs) and their corresponding success probabilities provided by the backward programming module, and combines them with real-time and predicted market data provided by the data fusion module to perform economic model calculations.

[0090] The strategy decision module includes a unit for calculating expected achievable revenue, a unit for calculating total execution cost, and a global optimal strategy solver. This strategy decision module provides solutions for each potential execution strategy. Calculate the Expected Strategy Profit (ESP). Execute the strategy. Defined as a binary tuple, by a specific dynamic processing chain and the planned start time Composition, that is .

[0091] The Expected Realizable Revenue Calculation Unit is responsible for estimating the total revenue that the strategy is expected to generate after considering all inherent risks and external market frictions. The calculation of this unit incorporates the probability of successful processing, the market value of the output, and its liquidity. The specific formula is as follows:

[0092] ;

[0093] in, Representation strategy The expected returns; This indicates the processing chain provided by the reverse planning module, which executes when processing a specific batch B of material. Probability of success; Summation symbol This indicates that the strategy iterates through and sums the results for each resource in the target product portfolio (TPP). This indicates the processing chain calculated based on the process knowledge base model, if the processing is successful. The first one that can be produced from material batch B target resources The quality; The estimated completion time of the strategy is equal to the start time. Add processing chain Expected execution time; This indicates that the system predicts the output time based on market data. ,resource Market unit price; This represents the market liquidity adjustment function, whose input is resources. At the moment of production Market liquidity indicators When market liquidity is excellent, the function value approaches 1; when liquidity is poor, i.e. there is a price but no market, the function value approaches 0, thus reducing the weight of this return.

[0094] The total execution cost calculation unit is responsible for calculating all costs required to implement the strategy. Its calculation comprehensively considers dynamically changing energy costs and time-related opportunity costs. The specific calculation formula is as follows:

[0095] ;

[0096] in, Representation strategy Total execution cost; Indicates the execution processing chain The instantaneous total power at time t; This indicates the real-time or predicted electricity price published by the power grid at time t; Indicates the execution processing chain Other operating costs that are not related to energy, such as fixed wear and tear on equipment and required labor costs; Indicates the time from the start of the strategy End time The cumulative energy cost throughout the process; This represents the storage and opportunity costs incurred by the delayed execution strategy; This refers to the current moment; if the strategy is to execute immediately... If the execution is delayed, this item will be zero. If the delay is selected, this item will include costs incurred due to material storage space occupancy and capital tie-up, depending on the length of the delay.

[0097] The global optimal policy solver is the final decision-making unit of this policy decision-making module. Within a preset future time horizon (e.g., the next 48 hours), the global optimal policy solver solves all candidate dynamic processing chains generated by the inverse programming module. And the start time of all discrete plans within the future time horizon. The resulting policy set is formed by combining policies. The global optimal policy solver will then iterate through each policy in the policy set. Calculate the corresponding expected strategy profit. Finally, by comparing the ESP values ​​of all policies, the global optimal policy solver determines and outputs the optimal policy with the maximum expected policy profit. And then transmit it to the scheduling and execution center.

[0098] The scheduling and execution center transforms the optimal execution strategy output by the strategy decision module into a set of timed operation instructions that the field industrial control system can execute, and ensures the closed-loop implementation of the decision.

[0099] The scheduling and execution center includes an automatic job instruction set generation unit, an interface unit for the field control system, and a forward-looking capacity management unit.

[0100] The job instruction set automatic generation unit is responsible for the optimal strategy received. The task instruction set automatic generation unit performs parsing and translation. It dynamically processes the chain. The process operation sequence included Decompose the sequence. For each process operation in the sequence... The unit retrieves the specific equipment identifiers (IDs) required to execute the operation, the key operating parameters that need to be set (e.g., target particle size of the crusher, magnetic field strength of the magnetic separator, air blowing pressure and delay of the classifier), and standard start-up and shutdown procedures from the process knowledge graph database. Finally, the unit generates a structured set of operation instructions. This instruction set not only includes the aforementioned equipment-level parameters and control commands, but also specifies the execution order, timestamps, and logical dependencies between each command, forming a complete automated production operation script.

[0101] The interface unit of the field control system is responsible for reliably transmitting the generated set of work instructions downwards and establishing a feedback loop. It communicates with the Manufacturing Execution System (MES) of the solid waste treatment plant or directly with the underlying Programmable Logic Controller (PLC) via Industrial Ethernet and standard industrial communication protocols such as OPC-UA. This communication occurs at the planned start time of the optimal strategy. Upon arrival, the commands in the work instruction set are issued one by one in a preset order. The interface unit of the field control system continuously receives status feedback from the field control system, such as whether the equipment has been successfully started, whether the parameters have been set in place, and whether the current process has been completed. This enables real-time monitoring of the execution status of the entire work process, ensuring that decisions are executed accurately.

[0102] The forward-looking capacity management unit transforms the "waiting time" in decision-making into "preparation time" for improving production efficiency. When the strategy decision-making module calculates the optimal strategy... Plan start time This forward-looking capacity management unit is activated at some point in the future to analyze the dynamic processing chain that will be executed. From the present moment to During this waiting period, a series of "preparatory" or "maintenance" instructions are proactively generated and issued to ensure that future tasks can be executed with the highest efficiency and success rate. Specific instruction types include:

[0103] Arrange preventative maintenance or calibration procedures for critical equipment that will be used soon (such as high-precision optical sorters);

[0104] Dispatch yard equipment to pre-move batches of target materials to be processed near the feed inlet of the processing line to shorten loading time;

[0105] The relevant storage areas are organized to reserve space for upcoming finished products. In this way, the forward-looking capacity management unit proactively optimizes the plant's future processing capacity as a manageable asset, thereby maximizing every profit opportunity created by the market.

[0106] The operations visualization module presents the complex data flow, model calculation process, and final decision results within the system to operations managers in an intuitive and understandable way, and provides an operational entry point for necessary manual supervision and intervention. This module includes a decision panoramic data display unit, a decision process traceability unit, and a manual intervention and strategy configuration unit.

[0107] The decision-making panoramic data display unit is responsible for building a comprehensive monitoring dashboard that graphically presents the real-time panoramic status of factory operations. Specifically, the interface of this decision-making panoramic data display unit will display:

[0108] Real-time time-of-use electricity price curves and market price fluctuation charts for various renewable resources are derived from the data fusion module;

[0109] It manages the real-time operating status, power load, and material inventory levels of each core piece of equipment on the production line.

[0110] The current optimal execution strategy determined by the strategy decision module The core summary information includes the strategy’s expected start and end times, the dynamic processing chain used, and its calculated expected strategy profit (ESP).

[0111] When operations and management personnel need to understand why the system made a specific decision, the decision process traceability unit provides a drill-down query function for decision details. Users can select the optimal strategy that has been executed or is yet to be executed on the interface. The decision process traceability unit retrieves and displays all the calculation basis for that strategy at the time of decision-making. The displayed content includes: a ranking list of all candidate strategies evaluated at that time and their ESP scores, allowing users to make horizontal comparisons; for the selected strategy, it lists in detail the numerical values ​​of each component of its expected achievable benefits and total execution costs. For example, it clearly shows the success probability, estimated output, market price, liquidity discount factor, and specific values ​​of energy cost credits, operating costs, and warehousing costs on which the calculation is based.

[0112] The manual intervention and strategy configuration unit provides a human-machine collaboration interface to ensure the robustness of the final decision. In the manual review mode, after the strategy decision module generates a new optimal strategy, the strategy is first sent to the manual intervention and strategy configuration unit and highlighted as "Pending Review" on the operation visualization module interface. Operation managers can review the details of the strategy within a preset time window, with clear operation controls allowing them to perform actions such as "Approve Execution," "Reject," or "Manual Selection." If "Manual Selection" is selected, the interface will display a list of other suboptimal candidate strategies, allowing personnel to choose different strategies to execute based on other non-model-based experience judgments. Furthermore, the manual intervention and strategy configuration unit also provides a backend configuration interface for authorized system administrators to adjust key parameters in the system model, such as the weight coefficients in the value liquidity scoring model. and , or the risk preference parameter in the expected strategy profit model.

[0113] See attached document Figure 2 , Figure 2 This is a schematic flowchart of a method according to an embodiment of the present invention. The present invention provides a method for energy consumption monitoring and resource utilization optimization of highway solid waste treatment equipment. This method can be executed by the aforementioned system and includes the following steps:

[0114] The S100 acquires and integrates multi-source heterogeneous data in real time, including internal equipment status data of the solid waste treatment production line, material flow data of batches of solid waste to be treated, and external energy and renewable resource market data. It parses the material flow data to generate probabilistic component vectors that quantify the uncertainty of the components, and performs standardization and timestamp alignment processing on all data to form a synchronous panoramic status data stream.

[0115] S200, based on market data from the panoramic status data stream, quantifies the market liquidity index of each potentially productive renewable resource and, combined with its market price, calculates its comprehensive score through a value liquidity-weighted scoring model. Based on the comprehensive score, it dynamically selects and generates a target product portfolio (TPP) for the current market environment.

[0116] S300 uses the target product combination (TPP) as the planning objective and performs a reverse path search in a pre-defined process knowledge graph to generate at least one technically feasible dynamic processing chain (DPC) for the target product combination. Furthermore, by incorporating probabilistic component vectors, it evaluates the probability of successfully executing each DPC to process a specified batch of solid waste.

[0117] S400, for each generated dynamic processing chain and its success probability, constructs a strategy set consisting of the dynamic processing chain and multiple candidate execution start times within a pre-defined future time horizon. For each strategy in the strategy set, calculates its expected achievable revenue and total execution cost to determine its expected strategy profit (ESP). Compares the expected strategy profits of all strategies and selects and determines the optimal execution strategy with the maximum expected strategy profit.

[0118] The S500 parses the optimal execution strategy into a structured set of work instructions containing equipment control parameters and execution timing. This set of instructions is then sent to the industrial control system at the solid waste treatment site via an industrial communication interface to drive the physical equipment. Simultaneously, it monitors the work execution status to form a closed-loop feedback loop, and performs proactive capacity management tasks when there is a waiting period for the optimal strategy.

[0119] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A system for monitoring energy consumption and optimizing resource utilization of highway solid waste treatment equipment, characterized in that, include: The data fusion module is used to acquire and fuse multi-source heterogeneous data in real time. The multi-source heterogeneous data includes solid waste treatment equipment status data, material flow data of batches of solid waste to be treated, and external market data. The module also parses the material flow data to generate a probabilistic component vector. The market insight module is used to quantitatively calculate the market liquidity index of each potential productive renewable resource based on the external market data provided by the data fusion module, and dynamically screen and generate target product combinations under the current market environment. The reverse planning module is used to perform reverse path search in a preset process knowledge graph library with the target product combination as the planning target, generate at least one dynamic processing chain for the target product combination, and evaluate the success probability of executing each dynamic processing chain to process a specified batch of solid waste in combination with the probabilistic component vector. The strategy decision module is used to calculate the expected strategy profit for each generated dynamic processing chain and its corresponding success probability, and to determine the optimal execution strategy with the maximum expected strategy profit. The scheduling and execution center is used to parse the optimal execution strategy into a structured set of operation instructions containing equipment control parameters and execution timing, and then send it to the industrial control system. The reverse planning module includes: The reverse path search unit is used to recursively trace back to the initial process node with raw solid waste as input from the final process operation node that can directly produce the target product in the process knowledge graph database, thereby generating a dynamic processing chain. The success probability assessment unit is used to calculate the success probability by comparing the minimum material requirements of the first process operation of the dynamic processing chain with the material characteristics of the batch of solid waste to be treated as described by the probabilistic component vector. The success probability assessment unit is specifically used for: From the probabilistic component vector, a probability distribution vector representing the availability of various related resources in the batch of solid waste to be treated is inferred. Each element in the probability distribution vector is a probability distribution used to describe the uncertainty of the availability of the corresponding resource. Extract a deterministic threshold vector of the minimum required quantity for the first process operation from the dynamic processing chain, where each element in the deterministic threshold vector is a definite value; The success probability is determined by calculating the probability that the available resource quantity described by the probability distribution vector is greater than or equal to the minimum required quantity defined by the deterministic threshold vector. The scheduling and execution center includes: The automatic job instruction set generation unit is used to decompose the dynamic processing chain in the optimal execution strategy into process operation sequences, and query the equipment identifier and key operating parameters for each process operation sequence according to the process knowledge graph library to generate a structured job instruction set. The interface unit of the field control system is used to send the structured set of operation instructions to the industrial control system when the planned start time of the optimal execution strategy arrives, and to receive execution status feedback for real-time monitoring. The forward-looking capacity management unit is used to proactively generate and issue preparatory or maintenance instructions related to the dynamic processing chain during the waiting period from the current time to the planned start time when the planned start time of the optimal execution strategy is later than the current time.

2. The energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment according to claim 1, characterized in that, The data fusion module includes: The data acquisition unit is used to acquire the status data of the solid waste treatment equipment, the material flow data of the batch of solid waste to be treated, and the external market data through industrial standard protocols, online rapid component analysis devices, and application programming interfaces, respectively. The probabilistic component vector generation unit is used to input the material flow data of the batch of solid waste to be processed acquired by the data acquisition unit into a pre-trained material composition inference model to output the probability distribution of the content of each recyclable resource, thereby generating a probabilistic component vector. The data preprocessing and synchronization unit is used to clean and normalize the solid waste treatment equipment status data, the material flow data of the batch of solid waste to be treated, and the external market data acquired by the data acquisition unit, and to assign a unified timestamp to align them in the time dimension, forming a synchronized time-series data stream.

3. The energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment according to claim 1, characterized in that, The market insight module includes: The market liquidity quantification unit is used to calculate the market liquidity index for each potentially productive renewable resource based on the trading volume, trading frequency and bid-ask spread in the external market data. The value liquidity weighted scoring unit is used to combine the market price of the renewable resources and the market liquidity index to calculate the value liquidity score of the renewable resources through a weighted summation model. The target product combination generation unit is used to determine and generate a target product combination from candidate resources based on the value mobility score and according to a preset screening logic.

4. The energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment according to claim 3, characterized in that, The formula for calculating the value liquidity score is as follows: ； in, It is a resource At any moment The overall score; It is a resource At any moment Market price after normalization; It is a resource At any moment Market liquidity indicators after normalization; and These are preset weighting coefficients.

5. The energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment according to claim 1, characterized in that, The strategy decision-making module includes: The expected achievable revenue calculation unit is used to integrate the success probability, the market value of the product, and market liquidity to estimate the expected achievable revenue for each potential execution strategy. The total execution cost calculation unit is used to comprehensively calculate the energy cost, operating cost, and warehousing and opportunity cost generated during the execution of each potential execution strategy in order to determine the total execution cost. A global optimal strategy solver is used to calculate the expected strategy profit by subtracting the total execution cost from the expected achievable profit, and to traverse all potential execution strategies to determine the optimal execution strategy with the maximum expected strategy profit.

6. The energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment according to claim 1, characterized in that, The system also includes an operations visualization module, which includes: The decision-making panoramic data display unit is used to graphically display the external market data, the solid waste treatment equipment status data, and the optimal execution strategy information in real time on the monitoring dashboard. The decision process traceability unit is used to respond to a query for a selected strategy, retrieve and display the calculation basis of the strategy during the decision-making process, the calculation basis including a sorted list of expected strategy profit scores of candidate strategies, and the component values ​​of the expected achievable revenue and total execution cost of the selected strategy. The manual intervention and strategy configuration unit is used to present the optimal execution strategy as pending review so that operations and management personnel can perform approval, rejection or manual selection operations.

7. A method for monitoring energy consumption and optimizing resource utilization of highway solid waste treatment equipment, applied to the energy consumption monitoring and resource utilization optimization system for highway solid waste treatment equipment as described in any one of claims 1-6, characterized in that, Includes the following steps: S100. Real-time acquisition and fusion of multi-source heterogeneous data, including solid waste treatment equipment status data, material flow data of batches of solid waste to be treated, and external market data, and parsing the material flow data to generate a probabilistic component vector; S200. Based on the external market data, quantitatively calculate the market liquidity index of each potential producible renewable resource, dynamically screen and generate target product combinations under the current market environment. S300. Taking the target product combination as the planning target, perform reverse path search in the preset process knowledge graph library to generate at least one dynamic processing chain for the target product combination, and combine the probabilistic component vector to evaluate the success probability of executing each dynamic processing chain to process the specified batch of solid waste. S400. For each generated dynamic processing chain and its corresponding success probability, calculate the expected strategy profit and determine the optimal execution strategy with the maximum expected strategy profit. S500: The optimal execution strategy is parsed into a structured set of operation instructions containing equipment control parameters and execution timing, and then sent to the industrial control system.

Images