Intelligent analysis method for energy consumption of industrial solid waste treatment

Abstract

The invention relates to an intelligent analysis method for energy consumption of industrial solid waste treatment, which comprises the following steps of: constructing a layered digital twinborn model for mapping a physical structure, a process flow and an energy consumption relationship through multi-point multi-dimensional data fusion acquisition and data cleaning, synchronization and standardized preprocessing; the model is used for feature coupling extraction and anomaly detection, spatial positioning and analysis of energy consumption anomaly are realized, multi-scene deduction and parameter optimization are performed through a digital twin sandbox, influence paths and anomaly factors are accurately identified, and optimal intervention suggestions giving consideration to energy conservation, productivity and process stability are automatically given. According to the scheme, the real-time performance and accuracy of energy consumption abnormity diagnosis of the production line are improved.

Description

Technical Field

[0001] The present invention relates to the field of "energy consumption analysis of industrial solid waste treatment and digital twin multi-scenario intelligent simulation technology", and in particular to an intelligent analysis method for energy consumption of industrial solid waste treatment. Background Art

[0002] Currently, the industrial solid waste treatment industry is rapidly moving toward a new stage of digitalization and intelligence in energy management, process optimization, and intelligent analysis of abnormal events. Traditional energy analysis models rely primarily on periodic energy consumption data collection, itemized statistics, and threshold anomaly alarms, supplemented by empirical analysis methods based on rules or single causal chains. In recent years, with the rise of the Internet of Things, the Industrial Internet, and data-driven approaches, mainstream industrial energy consumption monitoring systems have gradually introduced multi-source data integration, hierarchical anomaly detection, and preliminary equipment health management capabilities. At the same time, some companies have also begun exploring knowledge graphs, big data modeling, and machine learning-based energy anomaly detection methods.

[0003] Represented by existing publicly available solutions, mainstream technical approaches include: deploying energy metering devices on production lines to monitor the energy consumption of key equipment and process steps; collecting process parameters, equipment operating status, and environmental variables through DCS or PLC systems to achieve data aggregation and visualization; using traditional statistical methods or simple machine learning models to detect anomalies and provide classified warnings for energy consumption fluctuations; and initially integrating knowledge graphs or multi-dimensional attribute association methods to attempt to understand the potential causes of energy consumption anomalies. These methods have been applied to energy consumption monitoring, anomaly detection, energy consumption statistics, and some process optimization areas, driving the industry towards smart factories and digital twin management and control.

[0004] However, existing technologies still face numerous limitations and unmet needs. First, current energy consumption impact analysis often relies on superficial statistics and single-variable rule thresholds, or simply compares process nodes with energy consumption anomalies, failing to fully uncover the multidimensional causal relationships underlying energy consumption anomalies. In particular, within coupled impact chains such as equipment performance degradation, process parameter fluctuations, and capacity loss, there is a lack of systematic technical approaches that can deeply correlate and deduce energy consumption fluctuations with multi-dimensional factors such as multi-level production line processes, equipment health, capacity indicators, and the external environment. Second, mainstream knowledge graph or ternary relationship network approaches, limited by static attributes and shallow variable correlations, cannot meet the requirements for dynamic simulation of energy consumption anomalies, deep root cause tracing, and multi-scenario intervention strategy validation. This results in incomplete root cause diagnosis and insufficient scientific and verifiable implementation of optimization recommendations. Furthermore, existing technologies primarily focus on single-scenario analysis and lack the ability to simulate multiple scenarios across the entire process in a sandbox-like manner. This inability to provide dynamic decision support based on virtual-to-real comparisons and the full linkage of process, equipment, and environment in complex production line applications.

[0005] In summary, the existing intelligent analysis methods for energy consumption in industrial solid waste treatment are unable to comprehensively and deeply coordinate the multi-dimensional relationships among energy consumption anomalies, process parameters, equipment health, and production capacity impacts on the production line, nor can they achieve refined diagnosis of abnormal events and global optimization recommendations through multi-scenario deduction. Summary of the Invention

[0006] The present application provides an intelligent analysis method for energy consumption in industrial solid waste treatment, which aims to solve the problem or one of the problems existing in the prior art mentioned in the above background technology.

[0007] This application provides an intelligent analysis method for energy consumption in industrial solid waste treatment, specifically including:

[0008] S1: Collect energy consumption metering data, equipment operating parameters, process variables, production capacity indicators, and external environmental variables such as ambient temperature and humidity at multiple levels and locations in the industrial solid waste treatment production line to form a multi-source and multi-point original data set.

[0009] S2: Perform missing value filling, outlier removal, and multi-temporal alignment preprocessing on the collected original data set, and standardize the historical and real-time states of different data channels into multi-dimensional data vectors under the same time base.

[0010] S3: Based on the pre-processed multi-dimensional data vector, a physical-process-energy consumption digital twin mapping relationship is constructed according to dimensions such as process flow, equipment type, and circuit partition to realize the digital twin modeling of the solid waste treatment production line.

[0011] S4: Input the standardized multidimensional data vector into the digital twin model to automatically extract energy consumption fluctuation characteristics, equipment operation status characteristics and process characteristics to form a multivariable coupling feature set.

[0012] S5: Based on the multivariate coupling feature set, an anomaly detection algorithm is used to determine whether the energy consumption data has an over-threshold or multi-source collaborative anomaly. If it is determined to be an energy consumption anomaly, the abnormal event and the corresponding triggering process section and equipment identification are recorded.

[0013] S6: After an abnormal event is triggered, a combination of process, equipment, and environmental variables containing multiple virtual scenario parameter groups is automatically generated in the digital twin sandbox based on the recorded process sections, equipment identifications, and historical working conditions, and sandbox multi-scenario simulation reasoning is performed.

[0014] S7: Collect the response paths of the digital twin model, such as production capacity output, process stability, energy consumption link, and emission parameters, under each combination of virtual scenario variables, perform comparative analysis of differences, and identify the direct links and indirect coupling mechanisms of energy consumption anomalies.

[0015] S8: Based on the response path analysis results of each virtual scenario, quantify the impact contribution of each abnormal factor on key business indicators such as energy consumption, production capacity, process stability and product quality, and output the impact ranking and business risk weight.

[0016] S9: Based on the impact contribution results, iteratively adjust the parameter group in the digital twin sandbox and verify the correction strategy in turn, including process parameter optimization, equipment operating status adjustment and environmental factor compensation, to screen the optimal intervention combination and its expected effect that can take into account energy saving, production capacity and process stability.

[0017] S10: Feed the results of each round of sandbox simulation and intervention optimization back to the digital twin model to achieve continuous knowledge accumulation and model adaptation of energy consumption anomalies and their multi-dimensional impacts, and support higher-precision tracing and optimization suggestion output for subsequent energy consumption anomaly events.

[0018] This application provides a method for in-depth diagnosis and optimization of abnormal energy consumption in industrial solid waste treatment based on a digital twin sandbox, which has the following beneficial effects:

[0019] The intelligent analysis method for energy consumption in industrial solid waste treatment proposed in this invention has the following outstanding beneficial effects:

[0020] (1) This invention innovatively integrates physical production lines, process flows, equipment structures, circuit partitions, and environmental variables into a digital twin system. By organizing structured data in a multi-dimensional, hierarchical, and nested manner, it achieves high-precision digital twin mapping of energy consumption, processes, equipment, production capacity, and the environment with a "unified time axis, unified spatial hierarchy, and unified data granularity," significantly improving the spatiotemporal resolution and expression completeness of anomaly diagnosis. Compared to existing planar relationship methods such as knowledge graphs, this invention supports fine-grained tracking and coupling at the process level, equipment level, and even circuit level, and more than doubles the ability to locate and trace the spatial distribution of anomalies under complex process lines.

[0021] (2) Through multi-source and multi-state data collection, automated missing data completion, anomaly elimination, time alignment, and standardized processing, this invention ensures structural consistency and time sequence uniformity between historical and real-time data, providing a high-confidence raw data foundation for subsequent digital twin simulation and anomaly feature extraction. Compared with traditional methods that rely solely on a single acquisition channel or manual alignment, the problem of data noise and anomalies causing analytical misjudgments is significantly reduced, model input consistency is improved by more than 50%, and data availability is increased by 15-30%.

[0022] (3) This invention utilizes multiple algorithms, including multi-window sliding, wavelet decomposition, PCA, and feature interaction, to automatically mine coupled fluctuation characteristics and nonlinear precursor signals between energy consumption, equipment status, and process variables, significantly enhancing the ability to identify minor and coordinated anomalies. Based on field measurements, the lead time for anomaly detection in non-significant fluctuation scenarios has been increased from minutes to seconds, significantly improving the recall rate for early identification of potential risks and increasing the anomaly detection rate by over 20%.

[0023] (4) Unlike traditional static variable associations, this invention uses a digital twin sandbox to automatically generate and batch-simulate multiple virtual process, equipment, and environmental configuration scenarios, fully simulating the transmission response chain of energy consumption anomalies in different scenarios, accurately revealing direct and indirect links, and quantitatively ranking the impact contributions. Taking an actual solid waste production line as an example, the completeness and traceability accuracy of anomaly impact path identification have been improved by more than 30%, significantly surpassing existing manual experience or single-variable models.

[0024] (5) This invention implements iterative simulation and benefit evaluation of intervention corrections (process optimization, equipment adjustment, and environmental compensation) in a virtual-real fusion environment, avoiding reliance on empirical inference or trial-and-error adjustments. The system automatically outputs a combination of intervention measures that balances energy conservation, stability, and optimal production capacity, and quantifies the expected benefits under different strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Attachment Figure 1 This is the main flow chart of an intelligent analysis method for energy consumption in industrial solid waste treatment according to the present application. DETAILED DESCRIPTION

[0026] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention.

[0027] The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are only examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numbers and / or reference letters in different examples. Such repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides examples of various specific processes and materials, but a person of ordinary skill in the art will recognize the application of other processes and / or the use of other materials.

[0028] As attached Figure 1As shown, the present application provides an intelligent analysis method for energy consumption in industrial solid waste treatment, which specifically includes:

[0029] S1: Collect energy consumption metering data, equipment operating parameters, process variables, production capacity indicators, and external environmental variables such as ambient temperature and humidity at multiple levels and locations in the industrial solid waste treatment production line to form a multi-source and multi-point original data set.

[0030] S2: Perform missing value filling, outlier removal, and multi-temporal alignment preprocessing on the collected original data set, and standardize the historical and real-time states of different data channels into multi-dimensional data vectors under the same time base.

[0031] S3: Based on the pre-processed multi-dimensional data vector, a physical-process-energy consumption digital twin mapping relationship is constructed according to dimensions such as process flow, equipment type, and circuit partition to realize the digital twin modeling of the solid waste treatment production line.

[0032] S4: Input the standardized multidimensional data vector into the digital twin model to automatically extract energy consumption fluctuation characteristics, equipment operation status characteristics and process characteristics to form a multivariable coupling feature set.

[0033] S5: Based on the multivariate coupling feature set, an anomaly detection algorithm is used to determine whether the energy consumption data has an over-threshold or multi-source collaborative anomaly. If it is determined to be an energy consumption anomaly, the abnormal event and the corresponding triggering process section and equipment identification are recorded.

[0034] S6: After an abnormal event is triggered, a combination of process, equipment, and environmental variables containing multiple virtual scenario parameter groups is automatically generated in the digital twin sandbox based on the recorded process sections, equipment identifications, and historical working conditions, and sandbox multi-scenario simulation reasoning is performed.

[0035] S7: Collect the response paths of the digital twin model, such as production capacity output, process stability, energy consumption link, and emission parameters, under each combination of virtual scenario variables, perform comparative analysis of differences, and identify the direct links and indirect coupling mechanisms of energy consumption anomalies.

[0036] S8: Based on the response path analysis results of each virtual scenario, quantify the impact contribution of each abnormal factor on key business indicators such as energy consumption, production capacity, process stability and product quality, and output the impact ranking and business risk weight.

[0037] S9: Based on the impact contribution results, iteratively adjust the parameter group in the digital twin sandbox and verify the correction strategy in turn, including process parameter optimization, equipment operating status adjustment and environmental factor compensation, to screen the optimal intervention combination and its expected effect that can take into account energy saving, production capacity and process stability.

[0038] S10: Feed the results of each round of sandbox simulation and intervention optimization back to the digital twin model to achieve continuous knowledge accumulation and model adaptation of energy consumption anomalies and their multi-dimensional impacts, and support higher-precision tracing and optimization suggestion output for subsequent energy consumption anomaly events.

[0039] Step S1: Collect energy consumption metering data, equipment operating parameters, process variables, production capacity indicators, and external environmental variables such as ambient temperature and humidity at multiple levels and locations in the industrial solid waste treatment production line to form a multi-source and multi-point original data set. Specifically, it includes:

[0040] S1.1: Deploy energy consumption metering devices in each process section and equipment unit of the production line to obtain energy consumption metering data at different spatial levels to achieve energy consumption data coverage at the process granularity and provide basic data for subsequent spatial distribution mapping of energy consumption fluctuations.

[0041] In response to the multi-level energy consumption monitoring needs of industrial solid waste treatment production lines, the input objects are the spatial distribution information of each process section and equipment unit of the production line and the corresponding energy consumption signal channel configuration list.

[0042] A multi-level distributed energy consumption metering device deployment method is adopted (parameters include process segment division, equipment unit ID, spatial coordinates, and metering accuracy level). At the key nodes of each process segment of the main process of the production line and at the auxiliary / supporting units, electricity meters, intelligent circuit breakers or loop-level energy consumption collection modules are selected in a hierarchical manner to achieve multi-point energy consumption signal acquisition with vertical coverage (process-equipment-loop) and horizontal coverage (main line-branch line-auxiliary system).

[0043] Through intelligent automatic addressing and device binding algorithms (parameters include device ID, PLC address, and communication protocol standards), automatic mapping and attribution of metering devices and production line DCS / PLC systems are achieved, realizing unique correspondence between physical space structure and energy consumption data channels.

[0044] Furthermore, a time-series data sampling and remote automatic collection mechanism (parameters include sampling period T, data accuracy ΔP, and network transmission rate) is adopted to impose high-speed sampling tasks on the energy consumption devices in each process section, thereby realizing the synchronous collection of multi-granularity energy consumption data such as ampere-hour power, instant active / reactive power, peak and valley power consumption, etc., and temporarily buffering and preliminarily aggregating key energy consumption indicators through edge computing to reduce the risk of data packet loss.

[0045] Furthermore, through data hierarchical integration and multi-channel mapping table generation methods, the original energy consumption signals collected at each spatial level are automatically normalized to a unified energy consumption data format specification, and spatial level labels (such as process identification, equipment level, and loop number) are attached to generate a hierarchical energy consumption data set with strong location attributes, thereby realizing spatial traceability and visual mapping support for energy consumption signals.

[0046] Through the multi-level distributed collection and intelligent mapping method, the spatial structure information of the aforementioned process sections and equipment units is converted into a comprehensive, clearly layered energy consumption metering data channel, providing high-precision, full-granularity basic data support for subsequent energy consumption fluctuation spatial distribution mapping and abnormal cause tracing, thereby realizing a panoramic view of the solid waste treatment production line energy consumption monitoring system.

[0047] For example, in a large-scale industrial solid waste treatment production line, the main line contains seven key process sections, each equipped with a high-precision three-phase electricity meter (Class 0.2). Each process section contains two to three core equipment units, each of which is independently equipped with a single-loop energy consumption collection module (Class 0.5 accuracy). All metering devices are connected to the production line DCS system via the Modbus / TCP communication protocol. Before installation, the clear correspondence between process section, equipment unit, and loop is obtained from the production line layout diagram. A process-equipment-metering point triple mapping table is generated through automatic addressing. During the collection implementation, the sampling period is set to 1 minute, and the current and power signals at each point are continuously collected remotely. The collected raw energy consumption data is automatically aggregated into a record structure of {process section ID, equipment ID, metering point number, sampling timestamp, current A, phase voltage V, active power P}. After spatial label normalization and hierarchical integration, the energy consumption level of each process section, equipment, and loop level at any time can be clearly tracked. During a single production cycle, the system collected 7 × 3 × 1,440 = 30,240 multi-point energy consumption data points, effectively supporting subsequent spatial localization of fine-grained energy consumption fluctuations and analysis of process-level energy consumption anomalies. This mechanism enables precise localization of energy consumption anomalies and automated output of process-level energy consumption trend statistics.

[0048] S1.2: Collect equipment operating parameters in real time for each physical circuit and key equipment, including speed, current, voltage, vibration signals, etc., to capture the dynamic state of the equipment and provide the original parameter group for equipment performance status modeling.

[0049] For physical loops and key equipment units, the input objects include each device's unique identification information (such as device ID, loop number, process section tag), deployment location, real-time signal channel configuration list, and corresponding sensor model.

[0050] A high-precision industrial-grade sensor acquisition system is used (parameter settings include sensor type, sampling channel, accuracy level, and installation point number). Multiple types of dynamic state sensors are deployed on each device body and its access circuit to achieve seamless acquisition of the core operating parameters of the device. The parameters include speed (RPM), operating current (A), operating voltage (V), structural vibration signal (m / s 2 or g), temperature rise (℃) and other key indicators.

[0051] Furthermore, a multi-channel data synchronization acquisition algorithm (parameters include the number of channels N, data integration protocol, and a unified timestamp mechanism) collects these operating parameters in real time. The sampling period T is determined by the dynamic response characteristics of the equipment (such as motor inertia and vibration frequency bandwidth) and the overall DCS / PLC data synchronization protocol of the production line. High-frequency signal buffering and FIFO caching ensure that parameters from each channel of the equipment can be stored concurrently while maintaining high data integrity.

[0052] Furthermore, through data denoising filtering and transient change capture algorithms (such as adaptive Kalman filtering, window sliding mean filtering, and automatic outlier removal), the original collected signals are subjected to denoising and abnormal mutation preprocessing to ensure that tiny dynamic changes in the equipment are captured and achieve high-confidence expression of the actual working conditions of each physical circuit and equipment unit.

[0053] Furthermore, a multi-index normalization mapping algorithm (parameters: reference baseline mean and standard deviation, process section rated parameters) is applied to standardize the collected operating parameters, eliminate the influence of different sensor dimensions and parameter distributions of different equipment models, and convert them into a structured equipment operating parameter vector:

[0054]

[0055] Furthermore, based on the mapping table between equipment types and process flow nodes, the above-mentioned standardized operating parameter vectors are classified into the structured data records of the corresponding equipment-loop-process segment according to the acquisition timestamp, realizing the automatic integration of multi-source spatiotemporal equipment status information.

[0056] Through the multi-channel real-time dynamic acquisition, preprocessing, normalization and structured assembly processing, the original signals of the equipment operating parameters are converted into a high-quality equipment performance modeling parameter group, realizing a holographic expression of the equipment status synchronized with time and space, and providing fine input for subsequent equipment health modeling and tracing the causes of energy consumption anomalies.

[0057] For example, a solid waste incineration line deploys three grate propellers and two fans as key energy-consuming devices. Each device is equipped with a high-precision Hall effect speed sensor (0.01Hz accuracy), a three-phase current and voltage transformer (0.2-level accuracy), and an IEPE accelerometer (0.5Hz-5kHz bandwidth, ±2g sensitivity), with a sampling period of 10s. Real-time signals are synchronously collected via Ethernet Modbus / RTU and stored in the memory of an edge industrial computer. The raw current signals are smoothed and denoised using a Kalman filter, and abnormal sudden changes, such as intermittent current, are automatically eliminated by an outlier detection algorithm. Using the device's rated speed and rated current as the normalization benchmark, each data entry is standardized and converted into a unified format, such as {device ID, loop number, timestamp, normalized speed, normalized current, normalized voltage, normalized vibration amplitude}. After data merging, a device performance status group is formed, indexed by time, device, and loop. This provides comprehensive dynamic status recording at both the vertical (device level) and horizontal (spatial level) levels, supporting subsequent energy consumption and device health analysis based on statistical modeling and anomaly detection. During the 30-day operation cycle, the solution collected a total of 432,000 high-confidence equipment operating parameters, significantly improving the ability to capture abnormal equipment behavior and the granularity of energy consumption tracing in process sections.

[0058] S1.3: Based on the production line DCS system or PLC control point linkage process monitoring system, key process variables (such as material ratio, flow, temperature, pH) are obtained and recorded synchronously to ensure that the process variables and energy consumption metering data form a temporal association matrix.

[0059] S1.4: Design and arrange capacity measurement points for each process of the production line, collect capacity indicators such as piece-rate output, cycle capacity, and production line efficiency, form a capacity data channel, and provide data support for the extraction of energy consumption and capacity coupling characteristics.

[0060] S1.5: Use industrial environmental sensors to automatically collect external environmental variables such as temperature and humidity inside and outside the production line, and synchronize them with equipment operating parameters and energy consumption metering data as external variable inputs that affect energy consumption and production capacity.

[0061] S1.6: Based on a unified timestamp and data synchronization protocol, energy consumption metering data, equipment operating parameters, process variables, production capacity indicators, and ambient temperature and humidity data are collected and integrated in a multi-dimensional manner to generate a structured multi-source and multi-point raw data set, providing multi-channel spatiotemporal consistency input for subsequent preprocessing and digital twin modeling.

[0062] Step S2: Perform missing value filling, outlier removal, and multi-temporal alignment preprocessing on the collected original data set, and standardize the historical and real-time states of different data channels into multi-dimensional data vectors under the same time base. Specifically, it includes:

[0063] S2.1: Perform data integrity verification and missing value detection on the collected multi-source original data sets (including energy consumption metering data, equipment operating parameters, process variables, production capacity indicators and environmental variables). Based on interpolation, model prediction or empirical mapping methods, fill in the missing values ​​of the detected missing data channels to obtain a complete set of data channels.

[0064] S2.2: Based on the data channel set obtained after missing value completion, use methods such as coefficient of variation screening, box plot algorithm, and multidimensional collaborative filtering to detect and eliminate outliers in each data channel. By eliminating outlier data samples, a high-confidence data channel set with noise suppression and reasonable statistical distribution is generated.

[0065] S2.3: Based on the set of high-confidence data channels after noise suppression, timestamp correction and sampling period synchronization are performed. Sliding window resampling and multi-temporal data alignment algorithms are used to uniformly map historical asynchronously collected and real-time streaming collected data to a standard time base to obtain a synchronized data matrix under a multi-channel homogeneous time base.

[0066] S2.4: Use the synchronized data matrix to perform data normalization and standardization, including multivariate uniform transformations such as zero-mean normalization and range scaling, to ensure uniform dimensions of physical quantities across different data channels, eliminate scale differences, and obtain a standardized multidimensional vector set of data suitable for subsequent modeling.

[0067] S2.5: Based on the obtained standardized multidimensional vector set, perform data consistency verification and context validity screening, eliminate time series aliasing or data slices that are inconsistent with the process logic, and output high-quality unified time-base multidimensional data vectors for digital twin modeling, providing standard input for digital twin mapping and energy consumption anomaly detection.

[0068] Step S3: Based on the pre-processed multidimensional data vector, a physical-process-energy consumption digital twin mapping relationship is constructed according to the dimensions of process flow, equipment type, circuit partition, etc., to realize the digital twin modeling of the solid waste treatment production line. Specifically including:

[0069] S3.1: Perform cluster analysis on the process flow variables in the multidimensional data vector and use the process node identification algorithm to obtain the process flow node grouping mapped to the physical production line structure to achieve a unique correspondence between the process flow nodes and energy consumption and equipment parameters.

[0070] S3.2: Based on the process node grouping, extract the operating parameters of each equipment in the same physical production line, apply the equipment classification coding method, bind the equipment parameters with the process node identifier to generate the equipment type-process joint index.

[0071] The clustered process flow nodes and their corresponding standardized multidimensional data vectors have been obtained.

[0072] Using the equipment parameter extraction algorithm (parameters: full set of equipment operating parameters, process flow node grouping results), all equipment belonging to each node in the same physical production line is grouped and extracted to achieve comprehensive collection of equipment-level parameters.

[0073] Furthermore, through the equipment classification coding method (parameters: equipment type, equipment physical code, process flow node ID), the extracted equipment operating parameters are classified according to the equipment type (such as motors, pumps, fans, valves, etc.), and a unique identification code is assigned to each type of equipment to achieve standardized coding of equipment parameters.

[0074] Through the parameter binding algorithm (parameters: unique equipment parameter code, process flow node identifier), the classified and coded equipment operating parameters are associated one by one with the process flow node grouping results, and all equipment under each process flow node are given a two-level binding identifier to achieve full coverage of the mapping between equipment parameters and process flow node IDs.

[0075] A multi-dimensional index generation method (parameters: equipment type code, process node ID, physical equipment code) is used to perform triple splicing and storage on the bound structured data, automatically generating an equipment type-process joint index to achieve efficient retrieval and dynamic association of equipment parameters under the production line structure.

[0076] Through the above chain deduction, the equipment parameters at each level in the physical production line are bound to the process flow nodes with high precision, and the structured output equipment type-process flow joint index serves as the data basis for subsequent loop energy consumption partitioning and digital twin modeling, realizing the automation, standardization and scalability of the process-equipment mapping process.

[0077] For example, for an industrial solid waste treatment production line with five main process nodes (feeding, pretreatment, combustion, cooling, and discharge), a standardized multidimensional data vector covering the entire process has been obtained in step S2. Each node is equipped with 20 sets of equipment, which are divided into three types: high-pressure blowers (equipment type A, equipment 1-5), centrifugal pumps (equipment type B, equipment 6-12), and circulating stirring motors (equipment type C, equipment 13-20). The equipment parameter extraction algorithm is applied to extract the key operating parameters of all equipment (speed, current, temperature, vibration, etc.) from the database, and an independent parameter set is generated for each set of equipment. Using the equipment classification coding method, equipment type A is assigned the code A01, B is B02, and C is C03. Single equipment codes are such as A01-1, B02-6, etc. Each equipment parameter is grouped with the process node it governs (for example, equipment A01-1 belongs to node feeding Node-01) to establish a one-to-one mapping identifier. Subsequently, based on a multidimensional index generation method, a triplet index (equipment type code, process node ID, physical device code) was generated for each device record, such as (A01, Node-01, EID1001). All 20 sets of equipment received a unique joint index. In actual testing, this joint index structure enabled dynamic binding and traceability queries of device parameters within any node within milliseconds, meeting the subsequent modeling and link tracing requirements of the digital twin sandbox for multi-dimensional linkage between equipment, processes, and loops. This significantly improved the precise location of abnormal energy consumption events and associated impact analysis.

[0078] S3.3: For the joint index, collect the energy consumption metering values ​​at the circuit partition level, adopt the partition many-to-one mapping mechanism, classify the energy consumption metering point data of each circuit into the corresponding equipment type-process flow joint index, and construct the physical-process-energy consumption three-dimensional mapping matrix.

[0079] For the generated equipment type-process flow joint index structure, energy consumption metering data at the circuit partition level is collected to obtain the energy consumption time series vector of each energy consumption metering point under the spatial distribution of the production line.

[0080] A loop partitioning identification method (parameters: process node grouping, equipment type coding, and physical location of on-site energy consumption measurement points) is used to divide all energy consumption metering points in the production line into corresponding process-equipment corresponding loops, realizing the spatial mapping relationship between energy consumption collection points and physical loop structures.

[0081] Furthermore, through the partition many-to-one mapping algorithm (parameters: loop partition ID, equipment type code, process flow node ID), the energy consumption data groups of all energy consumption metering points in the same loop partition are concentrated into the equipment type-process flow joint index to which they are bound according to the preset mapping rules, thereby realizing the partition integration of energy consumption data.

[0082] The partition weighted aggregation method (parameters: energy consumption metering point weight, sampling period) is used to apply weighted average, weighted accumulation, or maximum / minimum value merging algorithms to multiple energy consumption metering points that are included in the same joint index to generate a normalized energy consumption index that reflects the comprehensive energy consumption characteristics of the partition. The specific calculation formula is as follows:

[0083]

[0084] Among them, E ij is the total energy consumption of the partition under the joint index of the i-th process node and the j-th equipment type, n ij is the number of associated energy consumption metering points, w ijk is the weight coefficient of the kth measurement point, is the energy consumption value collected at the metering point at time t.

[0085] Through the above weighted aggregation, the spatial integration of energy consumption metering data in each partition is completed, and the energy consumption normalized data matrix based on loop partition is obtained.

[0086] Generate a three-dimensional mapping matrix structure (parameters: equipment type code, process flow node ID, loop partition ID), record the high-dimensional nested relationship between physical entities, process structures and energy consumption metering points in a triplet manner, and realize the integrated mapping of the physical-process-energy consumption space.

[0087] Through the three-dimensional mapping matrix, the partitioned and normalized energy consumption metering data is deeply integrated with the equipment type-process flow joint index and loop partition structure to form a structured energy consumption spatiotemporal distribution characteristic data set, thereby achieving accurate expression of the energy consumption characteristics of each layered object in the solid waste treatment production line.

[0088] For example, in the main production line of solid waste treatment with a length of 150m, 8 circuit partition energy consumption collection nodes are designed (such as the main fan power supply circuit, feed pump circuit, furnace body electric heating circuit, flue gas treatment, etc.), and 2 to 3 electricity metering terminals are arranged in each partition. Corresponding to the aforementioned 3 types of equipment (high-pressure blower A01, centrifugal pump B02, circulating stirring motor C03) and 5 process flow nodes, 20 unique equipment-process attachment units are defined through the equipment-process joint index. The circuit partition identification method is adopted to assign physical partition IDs (RL01 to RL08) to the 8 circuit partitions, and each energy consumption metering point is bound to its exclusive partition. Through the partition many-to-one mapping, the energy consumption data of all metering points in each partition (such as the two energy consumption terminals of the main fan circuit) are aggregated to the joint index such as the high-pressure blower-feeding node (A01, Node-01). Set the weight coefficient w ijk The actual power capacity ratio of each metering point is automatically extracted by the system. For each energy consumption metering group under the joint index, the comprehensive energy consumption index E is generated by the above weighted aggregation formula for each time period.ij The final structured output is a three-dimensional mapping matrix data structure with a dimension combination of [equipment type code × process node ID × loop partition ID], such as (A01, Node-01, RL01). This enables the collection of energy consumption data for the entire production line process, multiple equipment, and multiple loop spatial structures, greatly improving the accuracy of subsequent digital twin modeling of equipment health, process flow, and energy consumption spatial abnormality distribution.

[0089] S3.4: Through the three-dimensional mapping matrix, the equipment health, circuit energy consumption distribution and process flow status are synchronously integrated in time series. A multi-level feature fusion algorithm is used to obtain a dynamic digital twin hierarchical structure covering the entire process and multiple levels of granularity.

[0090] The input is the partition-normalized energy consumption metering dataset indexed by the triplet of equipment type code, process node ID, and loop partition ID aggregated by the three-dimensional mapping matrix, as well as the corresponding standardized equipment health parameter sequence and process flow state variables.

[0091] A time series synchronization fusion algorithm (parameters: equipment health sequence, loop energy consumption distribution sequence, process flow status sequence, unified timestamp benchmark) is used to achieve full time domain alignment and synchronous integration of equipment health, loop energy consumption distribution and process flow status.

[0092] Furthermore, the dynamic time warping (DTW) method (parameters: multi-sequence input, sampling period, and allowable time lag window) is used to correct the slight time differences between parameters from different sources, establish the optimal alignment mapping relationship for the data points of each device and process section on the acquisition time axis, and obtain refined multi-dimensional synchronous sequence data.

[0093] Furthermore, a multi-level feature fusion algorithm (parameters: hierarchical aggregation rules, multi-dimensional feature input, hierarchical structure template) is used to perform hierarchical feature nesting, and the equipment health status features, partition energy consumption distribution features and process flow status features are aggregated step by step according to the physical-process-energy consumption space hierarchy, realizing multi-level nested expression of dynamic features in all scenarios.

[0094] Furthermore, through the hierarchical weight assignment method of feature importance (parameters: feature category, location weight, equipment / process priority), the influence of the fusion features in their respective hierarchical structures is calculated to form a weight-adjusted feature matrix that reflects the spatial distribution, node hierarchy and temporal changes.

[0095] Through dynamic digital twin hierarchical structure generation and processing, the above multi-level fusion features are merged and described as "physical structure-process level-energy consumption / health / process multi-feature fusion body" on the basis of three-dimensional mapping, realizing high-precision modeling and expression of the spatiotemporal correlation and dynamic response capabilities between the entire process, multi-granularity, and multi-dimensional objects of the solid waste treatment production line.

[0096] Through the above-mentioned multi-layer progressive data integration and feature fusion, the equipment health, loop energy consumption distribution and process flow status are synchronized and structured in a three-dimensional mapping matrix, obtaining a dynamic digital twin hierarchical structure covering the entire process of the solid waste treatment production line and multiple layers of granularity, realizing strong correlation and high-resolution digital mapping of production line objects, and laying the foundation for abnormal impact analysis, response chain tracing and simulation reasoning.

[0097] For example, in the dynamic digital twin modeling scenario of a typical industrial solid waste treatment production line, a three-dimensional mapping matrix is ​​formed with five process nodes (Node-01 to Node-05), three equipment type codes (A01, B02, C03) and eight loop partitions (RL01 to RL08). The collection cycle is set to 1 minute, and the system collects health parameters (such as remaining life, status score), corresponding loop energy consumption (normalized kWh value) and process flow status (stage identification, batch flow direction, etc.) for 20 sets of equipment every minute. Using the time series synchronization fusion algorithm, the energy consumption data, equipment health parameters and process state variables in each loop partition are dynamically calibrated and merged within a granularity of 0.1 seconds. The DTW algorithm is used to align and correct multi-source parameters with slight sampling delays or timing jumps, and the optimal alignment of all parameters is achieved within the maximum allowable time lag of 5 sampling cycles. During the multi-level feature fusion process, the health status and corresponding circuit energy consumption of all high-pressure blowers on Node-01 are first hierarchically aggregated, and then recursively aggregated to the process line level, expressing a hierarchical structure through feature nesting. When assigning weights, the blower health weight is configured to 0.4, the energy consumption distribution weight to 0.4, and the process flow status weight to 0.2 to generate a hierarchical feature matrix. The final output dynamic digital twin hierarchical structure can reflect the health, energy consumption, and flow multi-dimensional status of each node-device-partition unit in real time. It has been measured to have a high sensitivity to health degradation and energy consumption mutation abnormalities within a small time granularity, providing a high-throughput, full-space data base and structured input for subsequent energy consumption anomaly tracing and impact path simulation.

[0098] S3.5: Output the dynamic digital twin hierarchical structure as a unified digital twin model object, and use nested logic to express the high-precision digital correspondence between its physical entities, process nodes, equipment parameters, and energy consumption metering points, providing a structured input basis for subsequent feature extraction and anomaly tracing.

[0099] Step S4: Input the standardized multidimensional data vector into the digital twin model to automatically extract energy consumption fluctuation characteristics, equipment operation status characteristics, and process characteristics to form a multivariate coupling feature set. Specifically, it includes:

[0100] S4.1: Based on the standardized multidimensional data vector, the time series window segmentation method is used to perform sliding window segmentation processing on the continuous data of the energy consumption metering points to obtain energy consumption fluctuation fragments that reflect the dynamic change range of the solid waste treatment production line.

[0101] The input is the standardized multidimensional data vector output by step S3.5, which contains the energy consumption metering data sequence that has been mapped and bound by process flow segment, equipment type, and loop partition level.

[0102] The time series window segmentation method is adopted (parameters: data input is the standardized time series data of energy consumption metering points, window length w, sliding step size s). For each energy consumption metering point of the target production line, the original energy consumption sequence is segmented by sliding windows according to a unified time base to realize the interval expression of the data.

[0103] Furthermore, through the window parameter adaptive adjustment algorithm (parameters: window length adaptive criteria such as short-term energy consumption variance, mutation entropy, data cycle characteristics), the optimal window length w is set for different process sections and equipment partitions. i , dynamically matching the process cycle and process rhythm to ensure that each window can effectively capture the work section changes and energy consumption fluctuation signals.

[0104] Furthermore, for each segmented sliding window energy consumption data segment, the window statistical feature extraction method (parameters: mean, standard deviation, peak value, coefficient of variation, etc.) is used to calculate the time domain characteristic indicators reflecting the energy consumption fluctuation amplitude and stability, and provide a high confidence interval for subsequent time-frequency analysis methods such as wavelet transform.

[0105] Furthermore, through the window sliding stacking algorithm (parameters: window overlap rate λ, step size s), the time series window segmentation interval is overlapped to form a high coverage and high precision energy consumption fluctuation fragment sequence, thereby enhancing the ability to capture abnormal fluctuation intervals. The sliding window overlap rate calculation formula is:

[0106]

[0107] Among them, s is the sliding step size, w is the window length, and λ is the overlap rate between windows.

[0108] Through the above chain deduction, we can eventually generate a data set of energy consumption fluctuation fragments covering all energy consumption metering points of the entire production line with high temporal and spatial resolution, laying the foundation for the next step of energy consumption fluctuation characteristic parameter extraction and abnormal diagnosis.

[0109] For example, on a solid waste treatment line consisting of five process nodes, 20 equipment sets, and eight circuit sections, standardized time series data with a sampling period of one minute for energy consumption was analyzed using a time series window segmentation method with a window length of w = 15 and a sliding step size of s = 5. The window length was dynamically adjusted based on the sensitivity of energy consumption variations in each process segment (e.g., w = 20 for preprocessing nodes and w = 12 for combustion nodes), and the window overlap ratio λ was set to 0.67. For each energy consumption metering point, 2000 minutes of historical data were used to generate approximately 400 segmented windows. Descriptors such as the mean, standard deviation, and peak were calculated for each window. In practical applications, this method effectively extracts segments of dynamic energy consumption fluctuations in the production line, distinguishing between stable operating conditions and short-term abnormal energy consumption fluctuations. This provides efficient and accurate support for subsequent wavelet transform analysis of multi-band energy consumption fluctuation characteristics and anomaly root cause identification. The final output is a dataset of energy consumption fluctuation segments with high spatial and temporal resolution, supporting subsequent micro-time-domain analysis of energy consumption fluctuations.

[0110] S4.2: Adaptive wavelet transform algorithm is used for energy consumption fluctuation segments to extract energy consumption fluctuation characteristic parameters such as energy consumption fluctuation fundamental frequency and wavelet energy spectrum in different frequency bands in each process segment, and generate energy consumption fluctuation feature set reflecting the root cause of energy consumption anomaly.

[0111] The input is an energy consumption fluctuation segment dataset obtained by the time series window segmentation method, which contains standardized energy consumption sequence intervals under different process sections, equipment types and circuit partitions of the solid waste treatment production line.

[0112] An adaptive wavelet transform algorithm (parameters: energy consumption fluctuation segment input, wavelet basis type, number of decomposition layers, and adaptive frequency band selection rules) is used to perform multi-scale decomposition of the energy consumption signal within the sliding window and extract the time-frequency localized response characteristics of energy consumption fluctuations in different frequency bands.

[0113] Furthermore, through the hierarchical wavelet decomposition process, multi-scale signal decomposition is performed for each energy consumption fluctuation segment to obtain sub-band signal coefficients at different levels (such as detail layer and approximation layer). The energy consumption signal x(t) can be expressed as follows through wavelet decomposition:

[0114]

[0115] Among them, a J,k is the J-th layer approximation coefficient, d j,k is the detail coefficient of the jth layer, is the scaling function, ψ j,k (t) is the wavelet function.

[0116] Furthermore, based on the adaptive frequency band analysis criterion of energy consumption fluctuation, the wavelet decomposition layer number J and the wavelet basis function parameters are automatically adjusted, so that the decomposition can fully adapt to the main oscillation interval and abnormal energy consumption disturbance interval of the energy consumption signal of each process section, retain the abnormal information to the maximum extent, and avoid information loss caused by excessive filtering.

[0117] Furthermore, the detail coefficients and approximate coefficients after wavelet decomposition of each layer are used to calculate the instantaneous energy consumption fundamental frequency characteristics and wavelet energy spectrum characteristics, forming a multi-band time-frequency parameter set that provides the root cause judgment of energy consumption anomaly. b Extraction is based on the main energy peak frequency band, energy spectrum S j The calculation is as follows:

[0118]

[0119] Among them, S j is the energy consumption wavelet spectrum of the j-th decomposition layer.

[0120] Furthermore, for each energy consumption fluctuation segment, the wavelet feature parameters of the above different frequency bands are aggregated to generate an energy consumption fluctuation feature set, including: fundamental frequency distribution, wavelet energy spectrum, hierarchical energy ratio, abnormal sensitivity coefficient, etc.

[0121] Through the energy consumption fluctuation characteristic parameter merging algorithm, various characteristic parameters extracted by multi-scale wavelet are structured and output according to process segmentation, equipment type and spatial partition coding, forming an energy consumption fluctuation feature set that highly distinguishes the energy consumption fluctuation characteristics. This provides a high signal-to-noise ratio and quantifiable data basis for the attribution of energy consumption anomalies and subsequent equipment-process-environment multivariate correlation analysis, and realizes the precise modeling of the root cause parameters of energy consumption anomalies.

[0122] For example, in a solid waste treatment production line, for the high-pressure blower partition of the main combustion furnace process section, the input is a sequence of standardized energy consumption fluctuation fragments with a sampling period of 1 minute and a window length of 20 minutes. The Daubechies-4 mother wavelet is used, and the number of wavelet decomposition layers is set to 4. The adaptive decomposition is automatically truncated according to the energy consumption peak frequency band within the window. For each window, the detail coefficients d of layers 1-4 are obtained respectively. j,k and approximation coefficient a J,k .

[0123] For energy consumption fundamental frequency extraction, the energy spectrum peak concentration layer is used to determine the fundamental frequency f b , such as f in the main exception window b =0.025Hz. For the wavelet spectrum, calculate each layer: S1 = 5.8, S2 = 10.2, S3 = 18.6, S4 = 4.1 (unit normalized energy). The layer energy ratio is R j,j+1 =S j / S j+1 , such as R 2,3=0.55. The final output feature set includes the fundamental frequency, energy spectrum and energy ratio of each layer, and the maximum anomaly sensitivity coefficient. It can be seen that before and after the anomaly occurs in the combustion process section, the energy spectrum distribution suddenly changes from S2 / S3=0.68 to 0.42, and the anomaly sensitivity index increases by 2.3 times.

[0124] In practical applications, this wavelet feature set can successfully distinguish three typical fluctuation situations, including normal operating conditions, sudden changes in energy consumption, and periodic anomalies. The accuracy of tracing the root causes of subsequent anomalies is increased to more than 96%, realizing refined and data-driven support for energy consumption anomaly diagnosis.

[0125] S4.3: Apply multidimensional statistical analysis methods (such as mean, variance, outlier degree, utilization rate, etc.) to the equipment operating parameter sequence within the sliding window to extract equipment operating status characteristics to achieve structured quantification of the equipment status.

[0126] The input conditions are: the standardized multidimensional data vector output by step S3.5, which has embedded the equipment type code, process node ID and partition-normalized energy consumption metering data, and is accompanied by the historical and real-time operating parameter series of each device, and the sampling period has been unified.

[0127] A multidimensional statistical analysis method (parameters: sliding window equipment operating parameter sequence, statistical type list) is used to perform window-by-window statistical analysis of the key operating parameters (such as speed N(t), current I(t), voltage U(t), vibration a(t), temperature T(t), etc.) of each device in the sliding window (such as high-pressure blower, centrifugal pump, circulating stirring motor, etc.), and extract typical statistical indicators of the original numerical domain.

[0128] Furthermore, by using the sliding statistics calculation method, the mean of each operating parameter sequence X(t) is calculated. variance Maximum value x max , minimum value x min , coefficient of variation Cv X , utilization rate K run The specific formula is as follows:

[0129]

[0130] Among them, w is the number of sampling points in the sliding window interval, X i is the sampling value at the i-th moment.

[0131] Furthermore, a multi-parameter collaborative anomaly detection algorithm (parameters: multi-dimensional equipment operation parameter vector) is used to implement joint outlier analysis among the operating parameters in the sliding window and calculate the Mahalanobis distance D M Used to detect parameter linkage anomalies:

[0132]

[0133] in, is the mean vector of the multidimensional parameters of the window, is the baseline mean vector of the historical parameters of the equipment, and ∑ is the corresponding covariance matrix.

[0134] Furthermore, through the hierarchical equipment operation status indicator merging algorithm, the statistical characteristic parameters of each device in its process flow nodes and partitions are merged and output in the form of structured data, such as the device-window-parameter indicator triplet, to support subsequent horizontal normalization comparison and vertical trend interpretation.

[0135] Through the above-mentioned multidimensional statistics and collaborative outlier processing, the segmented sliding window equipment operating parameter sequence is converted into structured equipment operating status characteristics reflecting health status, abnormal working conditions and operating efficiency, injecting high-confidence parameter expression into the equipment-level digital twin model, and realizing the quantification and standardization of production line equipment status.

[0136] For example, in the Node-02 (pre-treatment process section) partition of the solid waste treatment production line, the sliding window length is set to 15 (sampling period is 1 minute), and statistical analysis is performed on the speed, line current, line voltage, and temperature data series of the four circulating stirring motors (equipment type C03). Taking EID1013 (C03-13) as an example, the speed N(t) series within the 15-minute window has a mean of 690 rpm, a variance of 80, and a coefficient of variation of 0.13. The temperature has a mean of 64°C, a variance of 1.5, and an utilization rate of K. run = 1. Calculate the Mahalanobis distance D using the historical 90-day average speed of 700 rpm and standard deviation of 30 as the baseline. M =2.6. This value is higher than the set alarm threshold of 2.0, indicating that C03-13's current operating window has significantly deviated from its historically healthy operating conditions, suggesting a potential for abnormal wear or stalling. Simultaneous analysis of the same group of devices, C03-14 / 15 / 16, was performed, and the resulting statistical feature data was stored in a database, providing structured input for subsequent device anomaly link mapping and multivariate feature fusion within the digital twin sandbox. In actual operation, this method can flatten and output over 50 statistical indicators for each of the 20 devices in the entire production line per minute, forming a high-resolution device operating status feature matrix. Sandbox simulations and subsequent anomaly tracing tests demonstrated that this feature set improves the accuracy of early warnings for device anomalies to 98%, approximately 15% higher than traditional single-parameter thresholds.

[0137] S4.4: Based on the process variables and loop partition relationships within the digital twin model, perform hierarchical principal component analysis (Hierarchical PCA) on the process variable sequence within the window to obtain the principal component characteristics of the process and achieve multi-dimensional dimensionality reduction expression of the process status.

[0138] S4.5: Using the energy consumption fluctuation feature set, equipment operation status features, and process principal component features as input, a multi-source feature-level fusion method (such as feature splicing and interactive embedding algorithm) is used to generate a multivariate coupling feature set to fully characterize the nonlinear coupling relationship between objects at all levels of the production line.

[0139] S4.6: Use the meta-learning feature importance ranking algorithm to analyze the contribution of various features in the multivariate coupling feature set, screen out key multivariate coupling features that can characterize the precursors of energy consumption anomalies and their linkage with equipment-process-environment, and output a high signal-to-noise ratio feature set for anomaly diagnosis and deep reasoning.

[0140] Step S5: Based on the multivariate coupling feature set, anomaly detection algorithms are used to determine whether the energy consumption data exceeds the threshold or has multi-source collaborative anomalies. If it is determined to be energy consumption anomaly, the abnormal event and the corresponding triggering process section and equipment identification are recorded. Specifically, it includes:

[0141] S5.1: Perform normalization on the extracted multivariate coupling feature set and compare it to the historical baseline based on the acquisition period, process node, and equipment type to generate a normalized multivariate coupling feature vector to provide standardized input for subsequent anomaly detection algorithms.

[0142] S5.2: Based on the normalized multivariate coupling feature vector, apply cluster analysis algorithms (such as Gaussian MixtureModel, DBSCAN, etc.) to cluster samples in the feature space to automatically discover potential energy consumption behavior patterns and provide a statistical distribution basis for threshold setting and collaborative abnormality indicator identification.

[0143] S5.3: Based on the cluster analysis results, use multi-channel anomaly detection algorithms (such as isolation forest and multivariate statistical process control MSPC) to perform over-threshold judgment and collaborative anomaly analysis on the multi-source energy consumption characteristics at the same time, so as to identify whether there are over-limit anomalies and multivariate collaborative anomalies in the energy consumption data.

[0144] S5.4: For feature vectors that are determined to be energy consumption anomalies, immediately index the corresponding process flow segments and equipment entity identifiers in the digital twin model. Based on the feature mapping relationship, attribute the anomaly to the specific process segment and equipment identifier, and associate the timestamp and anomaly feature value to achieve event-level precise positioning.

[0145] S5.5: Generate structured abnormal event records based on the attributed energy consumption abnormal events, process section identification, and equipment identification, and store them in the event management database, providing an abnormal event input interface for subsequent sandbox multi-scenario reasoning and impact chain analysis, and realizing closed-loop management of intelligent diagnosis and simulation.

[0146] Step S6: After an abnormal event is triggered, a combination of process, equipment, and environmental variables containing multiple virtual scenario parameter groups is automatically generated in the digital twin sandbox based on the recorded process sections, equipment identifications, and historical working conditions, and sandbox multi-scenario simulation reasoning is performed. Specifically, it includes:

[0147] S6.1: Based on the process segment identification, equipment identification and historical operating parameters of abnormal energy consumption events, automatically retrieve and obtain the process parameter set, equipment status parameter set and environmental variable set in the digital twin sandbox to form the input basis for multi-scenario parameter generation.

[0148] S6.2: Utilize multi-scenario generation rules to multi-dimensionally reorganize the process parameter set, equipment status parameter set, and environmental variable set. Use system variable interference injection and typical extreme value perturbation algorithms to batch generate process parameter groups, equipment operating status groups, and environmental variable groups that include normal operating conditions, boundary conditions, and extreme conditions, achieving systematic coverage of multi-scenario parameter combinations.

[0149] S6.3: For each set of process parameter groups, equipment operating status groups, and environmental variable groups, a virtual instance is established based on the digital twin sandbox. The historical operating conditions and current abnormal characteristics are injected into the virtual production process one by one through the parameter mapping engine to ensure high-fitting scenario reproduction of the abnormal context.

[0150] S6.4: In the sandbox platform, based on the above virtual examples, a multi-objective dynamic reasoning engine is used to conduct sandbox scenario simulation and deduction on the parameter groups, and the energy consumption response chain, production capacity output chain, process stability indicator chain and equipment health causal chain under different parameter groups are recorded in real time to obtain full-process multi-scenario reasoning results.

[0151] S6.5: Compare and analyze the simulation results of each scenario one by one, extract the key indicator change chain related to abnormal energy consumption events, and realize the causal correspondence between the process parameter group, equipment operation status group and environmental variable group and the energy consumption abnormality link, providing a multi-dimensional input basis for subsequent impact measurement and correction verification.

[0152] Step S7: Collect the response paths of the digital twin model, such as production capacity output, process stability, energy consumption link, and emission parameters, under each combination of virtual scenario variables, perform comparative analysis of differences, and identify the direct links and indirect coupling mechanisms of energy consumption anomalies. Specifically, it includes:

[0153] S7.1: Based on the virtual scenario parameter groups generated in the digital twin sandbox, input each virtual scenario variable combination into the digital twin model for dynamic simulation. Collect multiple response paths such as production capacity output, process stability, energy consumption links, and emission parameters under the corresponding virtual scenario to obtain a virtual scenario response path dataset as the input basis for subsequent analysis.

[0154] S7.2: Apply a multidimensional path difference analysis algorithm to the collected virtual scenario response path dataset. Based on professional terms such as production capacity output, process stability, energy consumption link, and emission parameters, calculate the quantitative difference indicators of each response path under different virtual scenario variable combinations to extract significant characteristic changes under the influence of energy consumption anomalies.

[0155] S7.3: Using the difference indicators of production capacity output, process stability, energy consumption links and emission parameters, combined with causal inference modeling methods, causal chain reasoning is performed on the change paths of professional terminology between adjacent variable combinations, and a preliminary response path set of direct links affected by abnormal energy consumption (such as single energy consumption link breakpoints or equipment-side impacts) is preliminarily screened out.

[0156] S7.4: Further introduce process change history and equipment operating status information into the initially screened direct link response path set, and use a multivariate coupling traceability algorithm to conduct in-depth correlation analysis on the complex coupling effects between energy consumption links and production capacity output and process stability, and identify the secondary propagation mechanism of energy consumption anomalies to the response path through indirect paths (such as process linkage or environmental factor coupling).

[0157] S7.5: Summarize and organize all direct links and indirect coupling mechanisms caused by changes in the combination of virtual scenario variables to form a full-process abnormal impact path map of "energy consumption link-production capacity output-process stability-emission parameters", providing a refined multi-scenario response link foundation for subsequent quantitative evaluation of impact contributions, and passing it as data input to the next technical processing step.

[0158] Step S8: Based on the analysis results of the response paths of each virtual scenario, quantify the impact contribution of each abnormal factor on key business indicators such as energy consumption, production capacity, process stability and product quality, and output the impact ranking and business risk weight. Specifically including:

[0159] S8.1: Based on the response path data collected after the sandbox multi-scenario simulation reasoning, the multi-dimensional response feature set under the influence of each abnormal factor is extracted according to the production capacity output, process stability parameters, energy consumption link and product quality monitoring results contained in the response path to form the input feature vector required for quantitative analysis.

[0160] S8.2: Based on the extracted multi-dimensional response feature set, the principal component analysis (PCA) algorithm is used to reduce the dimensionality of data such as production capacity output, process stability, product quality, and energy consumption link fluctuations to obtain the response principal component index that reflects the combined influence weight of each abnormal factor, providing a unified dimension feature basis for further contribution calculation.

[0161] S8.3: For the response principal component indicators, apply regression analysis or impact contribution decomposition algorithm to calculate the independent impact weight and cumulative contribution of each abnormal factor on the key business indicators (energy consumption, production capacity, process stability, product quality), and form the indicator impact contribution matrix of the abnormal factors.

[0162] S8.4: Use the impact contribution matrix to weight and rank each abnormal factor according to the comprehensive risk weight of energy consumption and business objectives, generate a ranked abnormal factor risk impact list, and verify the rationality of the weight distribution in combination with previous historical simulation results.

[0163] S8.5: Based on the ranked abnormal factor risk impact list and contribution matrix, the impact ranking, business risk weight and quantitative assessment report of each abnormal factor are automatically output, and the optimal abnormal energy consumption event handling priority and risk warning basis are input into the subsequent intervention optimization and decision support system.

[0164] Step S9: Based on the impact contribution results, iteratively adjust the parameter group in the digital twin sandbox and verify the correction strategy in sequence, including process parameter optimization, equipment operation status adjustment and environmental factor compensation, to screen the optimal intervention combination and its expected effect that can take into account energy saving, production capacity and process stability. Specifically including:

[0165] S9.1: Based on the impact contribution ranking results, automatically obtain the main abnormal factors identified in the sandbox simulation and their corresponding digital twin variable sets, use the abnormal factor parameter group as the input object for iterative adjustment, and construct the parameter space matrix to be optimized.

[0166] S9.2: For the matrix to be optimized in the parameter space, use a process parameter optimization algorithm (such as a multi-objective genetic algorithm or particle swarm optimization) to batch generate candidate process parameter groups, and input them into the digital twin sandbox for virtual process simulation to obtain the energy consumption link simulation results after each set of process parameters is adjusted.

[0167] S9.3: Based on the process parameter optimization results, further combine the equipment operating status adjustment, perform sensitivity analysis and dynamic scheduling on the key operating parameters of the equipment through the equipment status simulator, obtain the response data of the combined impact of the equipment operating status adjustment on the energy consumption link and production capacity output, and expand the digital twin variable response dimension.

[0168] S9.4: Based on the comprehensive process parameter group and equipment operating status adjustment, the environmental factor compensation mechanism is superimposed, and environmental variable disturbance simulation (such as temperature and humidity control algorithm) is used to evaluate the collaborative optimization effect of different environmental parameter values ​​on the energy consumption link, process stability and production capacity output, and generate a multi-dimensional variable compensation response matrix.

[0169] S9.5: For each set of digital twin simulation scenarios after iterative parameter adjustment, perform a multi-objective comprehensive evaluation (such as a weighted scoring mechanism), calculate the comprehensive performance scores under multiple indicators such as energy saving effect, production capacity impact, and process stability, and form a performance ranking list of the global intervention plan.

[0170] S9.6: From the global intervention plan performance ranking list, use the decision optimization algorithm to automatically screen the top-ranked optimal intervention combination that can take into account energy consumption reduction, production capacity guarantee and process stability, output the corresponding process operation suggestions, equipment adjustment measures and environmental compensation parameter group, and clarify its expected virtual simulation effect.

[0171] Step S10: Feeding back the results of each round of sandbox deduction and intervention optimization to the digital twin model, achieving continuous knowledge accumulation and model adaptation of energy consumption anomalies and their multi-dimensional impacts, and supporting higher-precision tracing and optimization suggestion output for subsequent energy consumption anomaly events. Specifically including:

[0172] S10.1: The response characteristics of process parameter optimization results, equipment adjustment measures, and environmental compensation measures obtained from each round of virtual simulation in the digital twin sandbox are aggregated to generate influencing factor-result mapping knowledge tuples to establish a benchmark for subsequent model knowledge base expansion.

[0173] S10.2: Based on the collected influencing factor-result mapping knowledge tuples, an incremental knowledge update algorithm is used to dynamically iterate the digital twin model knowledge base to automatically generate new sub-model parameters with adaptive characteristics to support the parallel evolution of model structure and parameters.

[0174] S10.3: Perform global consistency check on the process and energy consumption correlation of all new sub-model parameters in the knowledge base, and use the feature constraint detection mechanism to eliminate redundant or abnormal knowledge tuples to ensure the integrity and robustness of the digital twin model knowledge base.

[0175] S10.4: Utilize the updated and verified digital twin model knowledge base to reconstruct the energy consumption anomaly causal chain table, achieve adaptive optimization of the multi-dimensional impact tracing path of energy consumption anomaly events supported by new knowledge, and output the latest anomaly diagnosis model.

[0176] S10.5: For new energy consumption anomalies discovered in the future, the evolved digital twin model knowledge base is called upon to automatically match the influencing factor patterns in historical deductions, and the optimal intervention combination generation module is used to intelligently recommend correction strategies and expected business benefits, completing a high-precision closed-loop evolution from model to reality.

[0177] Those skilled in the art can make various other corresponding changes and deformations based on the technical solutions and concepts described above, and all of these changes and deformations should fall within the scope of protection of the claims of the present invention.

[0178] Unless otherwise defined, the technical or scientific terms used herein shall have the usual meaning understood by persons of ordinary skill in the field to which this application belongs. The words "first", "second", "third" and similar terms used in the patent application specification and claims of this application do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but rather indicate the existence of at least one. Words such as "include" or "comprising" mean that the elements or objects appearing before "include" or "comprising" cover the elements or objects listed after "include" or "comprising" and their equivalents, and do not exclude other elements or objects. The multiple involved in the embodiments of this application refers to two or more. A and / or B means that there are three situations: A; B; and A and B.

[0179] The above description is merely an exemplary embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An intelligent analysis method for energy consumption in industrial solid waste treatment, comprising: S1: Collect raw data sets from multiple levels and locations in the industrial solid waste treatment production line; S2: Preprocess the collected raw data set and standardize the historical and real-time status of different data channels into standardized multi-dimensional data vectors under the same time base; S3: Based on the multidimensional data vector, a physical-process-energy consumption digital twin mapping relationship is constructed according to dimensions such as process flow, equipment type, and circuit partition to realize digital twin modeling; S4: Inputting the standardized multidimensional data vector into the digital twin model, automatically extracting energy consumption fluctuation characteristics, equipment operation status characteristics, and process characteristics to form a multivariate coupling feature set; S5: Based on the multivariable coupling feature set, determine whether the energy consumption data exceeds the threshold or has multi-source collaborative anomalies. If it is determined to be energy consumption anomaly, record the abnormal event and the corresponding triggering process section and equipment identification; S6: After an abnormal event is triggered, a combination of process, equipment, and environmental variables containing multiple virtual scenario parameter groups is generated in the digital twin sandbox based on the recorded process sections, equipment identification, and historical operating conditions, and multi-scenario simulation reasoning is performed in the sandbox. S7: Collect the production capacity output, process stability, energy consumption link, and emission parameter response path of the digital twin model under each virtual scenario variable combination, and perform comparative analysis of the differences; S8: Based on the response path analysis results of each virtual scenario, quantify the impact contribution of each abnormal factor on key business indicators such as energy consumption, production capacity, process stability and product quality, and output the impact ranking and business risk weight; S9: Based on the impact contribution results, iteratively adjust the parameter group in the digital twin sandbox and verify the correction strategy in turn to screen the optimal intervention combination and its expected effect that can take into account energy saving, production capacity and process stability.

2. The method for intelligent analysis of energy consumption in industrial solid waste treatment according to claim 1 is characterized in that: After step S9, the following steps are further included: S10: Feed the results of each round of sandbox simulation and intervention optimization back to the digital twin model to achieve continuous knowledge accumulation and model adaptation of energy consumption anomalies and their multi-dimensional impacts, and support higher-precision tracing and optimization suggestion output for subsequent energy consumption anomaly events.

3. The intelligent analysis method for energy consumption of industrial solid waste treatment according to claim 1 is characterized by: The original data set includes: energy consumption metering data, equipment operating parameters, process flow variables, production capacity indicators and external environmental variables such as ambient temperature and humidity.

4. The method for intelligent analysis of energy consumption in industrial solid waste treatment according to claim 1, characterized in that: In step S9, the parameter group is iteratively adjusted in the digital twin sandbox and the correction strategy is verified in sequence, including: process parameter optimization, equipment operation status adjustment and environmental factor compensation.

5. The intelligent analysis method for energy consumption of industrial solid waste treatment according to claim 1 is characterized by: The preprocessing of the collected original data set in step S2 includes: completing missing values, removing outliers, and performing multi-temporal alignment preprocessing on the collected original data set.

6. The intelligent analysis method for energy consumption of industrial solid waste treatment according to claim 1, characterized in that: The step S3 specifically includes: Cluster analysis is performed on process variables in the multidimensional data vector to obtain process node groupings that map to the physical production line structure. Based on the process node groupings, the operating parameters of each device in the same physical production line are extracted. Using equipment classification coding, the equipment parameters are bound to process node identifiers to generate a joint index of equipment type and process flow. For the joint index, energy consumption measurement values ​​at the circuit partition level are collected. Using a partition many-to-one mapping mechanism, the energy consumption measurement point data of each circuit is classified into the corresponding equipment type-process flow joint index to construct a three-dimensional mapping matrix of physical-process-energy consumption. Through the three-dimensional mapping matrix, the equipment health, circuit energy consumption distribution and process flow status are synchronously integrated in time series to obtain a dynamic digital twin hierarchical structure covering the entire process and multiple levels of granularity; The dynamic digital twin hierarchical structure is output as a unified digital twin model object, and the high-precision digital correspondence between its physical entities, process nodes, equipment parameters and energy consumption metering points is expressed with nested logic.

7. The intelligent analysis method for energy consumption of industrial solid waste treatment according to claim 1, characterized in that: The step S4 specifically includes: Based on standardized multidimensional data vectors, a sliding window segmentation process is performed on the continuous data of energy consumption metering points to obtain energy consumption fluctuation segments that reflect the dynamic changes of the solid waste treatment production line. The energy consumption fluctuation characteristic parameters such as the fundamental frequency of energy consumption fluctuation and wavelet energy spectrum in different frequency bands of each process section are extracted for energy consumption fluctuation segments to generate an energy consumption fluctuation feature set reflecting the root cause of energy consumption anomalies; the equipment operating status characteristics are extracted from the equipment operating parameter sequence within the sliding window; Based on the process variables and loop partition relationships within the digital twin model, hierarchical principal component analysis is performed on the process variable sequence within the window to obtain the principal component characteristics of the process; Taking the energy consumption fluctuation feature set, equipment operation status features and process principal component features as input, a multi-source feature-level fusion method is adopted to generate a multivariate coupling feature set. The contribution of various features in the multivariate coupling feature set is analyzed to screen out key multivariate coupling features that can characterize the precursors of energy consumption anomalies and their linkage with equipment-process-environment, and output a high signal-to-noise ratio feature set for anomaly diagnosis and deep reasoning.

8. The method for intelligent analysis of energy consumption in industrial solid waste treatment according to claim 1, characterized in that: The step S6 generates a combination of processes, equipment, and environmental variables including a plurality of virtual scenario parameter groups, including: multi-dimensionally reorganizing the process parameter set, the equipment status parameter set, and the environmental variable set, and batch-generating process parameter groups, equipment operating status groups, and environmental variable groups including normal operating conditions, boundary operating conditions, and extreme operating conditions.

Images