126 results about "Quantile regression" patented technology

The invention discloses a box-type substation state monitoring and early warning method based on artificial intelligence, relates to the technical field of intelligent power grids, and aims to solve the problems of missing report, false report and response lag caused by the fact that an existing static threshold ignores multi-physical coupling and a depth model highly depends on scarce fault samples. According to the scheme, sliding window kernel density estimation is carried out on a multi-channel time sequence signal, a dynamic coupling matrix is constructed through recursion Copula decomposition, a three-level threshold surface is generated through time-varying quantile regression, abnormal samples and graph attention network extraction state representation are generated in combination with a conditional variation auto-encoder, lightweight recursion pruning is carried out, and the dynamic coupling matrix is obtained. An abnormal score is generated through a multilayer Bayesian network and particle filtering, a multi-step risk trend is discriminated through a Gaussian kernel derivative slope, and finally unscented Kalman filtering is used for smoothing and online threshold correction; according to the method, the detection sensitivity and the early warning recall rate of the box-type substation to the transient coupling fault are remarkably improved, the response speed is improved, and the false alarm frequency is effectively reduced.

The invention discloses an intelligent recommendation method for optimizing advertisement keyword combination through cross validation, and relates to the technical field of advertisement technology and search engine marketing, which comprises the following steps: constructing a heterogeneous data set through multi-modal data fusion, and layering according to data sparseness: training a Transform-XL time sequence model by adopting time cross validation of a dynamic K value in a high resource layer; a graph neural network association graph is introduced into a low resource layer, semantic expression of a long tail word is enhanced, a stratified sampling-transfer learning two-channel mechanism is designed, and the generalization ability is improved in combination with exposure frequency weighting and a parameter freezing strategy; developing a Bayesian fusion engine, and dynamically weighting a high / low resource layer prediction result by using an improved Materon kernel function Gaussian process; and generating a confidence interval based on neural quantile regression, and outputting an optimal keyword combination sequence under ROI-risk-diversity constraint in combination with multi-target Pareto optimization. According to the method, the cold start efficiency and the long-tail resource utilization rate are improved, and high-robustness decision support is provided for advertisement putting.

The invention discloses an industrial and commercial energy storage EMS scheduling system based on multi-objective optimization, and relates to the field of intelligent control of energy storage systems. The method is used for solving the problems of battery life attenuation, electricity price fluctuation and insufficient load abrupt change adaptability in energy storage scheduling. The service life management module analyzes the ion migration state to generate a dynamic power safety envelope line as a layered optimization hard constraint; the feature extraction module constructs an electricity price association network, predicts photovoltaic / load in combination with quantile regression and extracts net load abrupt change features; the hierarchical optimization module compresses the discharge depth and activates load regulation in demand risk, and synchronously optimizes economy, life loss and power grid constraint through taboo search multi-objective evolution; the execution arbitration module monitors the distance between the power and the envelope line in real time, and when the distance is lower than a threshold value, rolling optimization is started to improve the life weight and feed back the lithium precipitation characteristic to drive the envelope line to update, and closed-loop protection is formed.

The invention provides a construction method of a prediction model based on dynamic gating and cross-modal attention fusion, and belongs to the technical field of model construction. Comprising the following steps: constructing a feature coding layer, and respectively coding multi-source input data to obtain each modal feature vector; constructing a multi-source data interaction layer, and performing deep interaction on each modal feature vector; and finally, carrying out weighted fusion on the main modal features and the cross-modal interaction features based on a dynamic gating mechanism. Constructing a feature fusion layer, and performing time sequence pooling and full-connection fusion on the interacted multi-modal features to obtain a fusion feature vector; and constructing a quantile regression layer, and outputting a prediction result based on the fusion feature vector. According to the method, the prediction model is constructed by fusing the dynamic gating mechanism and the cross-modal attention mechanism, so that the problems of an existing prediction model in the aspects of deep fusion of multi-source heterogeneous data, cross-modal dynamic interaction and accurate quantification of tail risks are solved, and then efficient prediction of stock price collapse risks is realized.

The invention relates to a cultural scenic spot people flow prediction and scheduling method based on machine learning, and belongs to the field of scenic spot management. 2, constructing and training a mixed attention model, wherein the constructed mixed attention model comprises a spatial-temporal feature extraction module and a mixed attention mechanism module; step 3, people flow prediction and result output: inputting real-time data into the trained mixed attention model to carry out people flow prediction, and outputting a future time period passenger flow volume value; a quantile regression technology is combined to output a confidence interval of the passenger flow volume; step 4, scheduling strategy implementation: constructing a dynamic thermodynamic diagram based on a prediction result, and marking crowd density levels of different areas in the thermodynamic diagram; early warning information is pushed in real time, and touring suggestions are sent to tourists in advance.

The invention discloses a power system operation reserve quantification method, system and device based on photovoltaic probability prediction and a medium. The method comprises the following steps: calculating an Euclidean distance between a photovoltaic predicted value of a point to be decided and a historical photovoltaic predicted value, and searching a photovoltaic power generation historical data set similar to the point to be decided; performing quantile regression based on the similar historical data set to obtain quantiles corresponding to a plurality of tail end quantile levels, fitting a probability density function of photovoltaic prediction deviation by using a Gaussian mixture model, and further calculating a risk loss expectation of a point to be decided; and constructing a standby cost function, considering the reliability constraint of the standby, taking the sum of the minimum standby reserved cost and the risk loss expected cost as a target function, and finally optimizing to obtain the standby capacity of the power system. According to the method, the change of the reserve price along with the capacity is considered, the photovoltaic probability prediction information can be fully utilized, and a more economical and reliable power system reserve quantification result is obtained.

The invention belongs to the technical field of livestock and poultry breeding environment intelligent control, and particularly relates to a meat duck breeding environment temperature regulation and control method and system integrating quantile regression prediction and reinforcement learning. The method comprises the following steps: collecting multi-dimensional data of a breeding environment and operation state information of environment regulation and control equipment, and performing feature construction to obtain a feature vector; using the multi-source data set to predict in-house temperature based on quantile regression to obtain predicted temperatures under different quantiles; modeling a breeding environment temperature regulation problem into a Markov decision process, and dynamically adjusting fan power and a wet curtain equipment state; and setting an online updating model step threshold, and performing actual deployment and application on the reinforcement learning decision model obtained by training and fusing quantile regression information. According to the method, the problems that the meat duck breeding environment temperature regulation and control technology still faces insufficient predictability, regulation and control lag, control strategy static performance, multi-target optimization deficiency and the like are solved.

The invention discloses a photovoltaic power interval prediction method based on a GRU-LSTM combined neural network, and belongs to the technical field of photovoltaic power interval prediction. The method comprises the following steps: S1, taking historical power generation data as original wind-solar power generation power prediction data, processing the data, and screening related meteorological characteristics by adopting a Pearson correlation coefficient; s2, a GRU-LSTM combination model is constructed, and related hyper-parameters are set; s3, taking the screened related meteorological features as input for training, calculating a photovoltaic point prediction result according to a weight coefficient, and performing related error evaluation; and S4, based on the photovoltaic power point prediction result, calculating a photovoltaic power interval prediction result by using a quantile regression technology, and detecting performance evaluation through a test set. According to the method, the minimum prediction error correlation index is taken as the target, the influence of different weathers on photovoltaic power processing is considered, the Pearson's correlation coefficient analysis is utilized to select more representative meteorological characteristics, and the combined model and the quantile regression technology are utilized to finally obtain the photovoltaic power interval prediction result.

The invention relates to the technical field of power system operation and control, and particularly discloses a time-sharing electric quantity prediction method based on a logarithmic load density growth curve, which comprises the following steps of: firstly, performing causal detection and dynamic time-delay optimization on historical load and multivariate external data through convergence cross mapping and mutual information technologies, and constructing a causal time-delay feature set; and the problems of multi-element coupling and time-delay effect quantization are solved. Secondly, fitting a load trend by using time-frequency decomposition in cooperation with a segmented logistic model, extracting dynamic parameters representing a growth rate and a saturation capacity, and endowing the model with a sensing ability for a load evolution stage; then, causal features, growth parameters and load components are deeply fused through cross-domain modulation and a gating mechanism, the nonlinear modulation effect of an external environment on a load mode is explicitly modeled, and finally, a probability interval is generated in combination with quantile regression and residual error correction. According to the scheme, accurate and probabilistic prediction of the time-sharing electric quantity in a complex scene is realized, and the scientificity of an agent electricity purchase decision is improved.

The invention discloses a power distribution network photovoltaic openable capacity dynamic evaluation method fusing voltage stability margin and neural network optimization, and aims to solve the problems that a traditional method does not fully consider dynamic stability and is low in calculation efficiency. Firstly, a prediction method combining kernel density estimation and quantile regression is adopted to accurately quantify the uncertainty of distributed photovoltaic output. The core innovation of the method is that on the basis of traditional security constraints, a static voltage stability margin (VDSM) is introduced as a key constraint condition, and a double-layer interval analysis model capable of guaranteeing the dynamic stability of a power grid is constructed. Secondly, in order to efficiently solve, the invention provides a framework of'neural network pre-screening + parallel optimization ': after a model is decomposed into optimistic sub-problems and pessimistic sub-problems through an interval decoupling technology, massive candidate solutions are quickly screened by utilizing a neural network model, so that a feasible solution space is greatly reduced, and the solution efficiency is improved; and carrying out parallel optimization solution on the sub-models in combination with an improved particle swarm optimization algorithm.

The invention discloses a vehicle-mounted power supply short-circuit heat prediction method based on big data analysis, and the method comprises the following steps: collecting real-time operation data, and storing the data in a data cache, thereby obtaining a cache time sequence signal; filtering the cached time sequence signal, and extracting a basic feature sequence; constructing an RC network model based on the basic feature sequence, and converting a fractional order operator into an equivalent integer order model; performing parameter estimation on the equivalent integer order model to generate a thermal prediction model; taking the thermal prediction model and the basic feature sequence as joint input and output power loss features; inputting the power loss characteristics into a quantile regression forest model to form a heat prediction result; comparing the heat prediction result with a heat threshold value and a temperature rise threshold value, and outputting a control trigger signal; and calling parameters in a control strategy library according to the control trigger signal to generate a control strategy. According to the invention, vehicle-mounted power supply short-circuit heat prediction is realized.

The invention belongs to the technical field of energy system planning, and particularly relates to a park integrated energy system stochastic planning method and system based on multiple uncertainties, and the planning method comprises the steps: building a probability model of a multi-energy load growth rate based on park industrial planning and historical data; utilizing Monte Carlo simulation and K-means clustering to generate a representative load scene tree; establishing an upper and lower boundary prediction model of the energy price and the equipment cost by adopting a quantile regression forest method; constructing a multi-stage collaborative optimization model taking the minimum comprehensive cost expectation as a target, and considering constraint conditions such as power flow, operation, time sequence and space; and carrying out reverse recursion solution by utilizing a dynamic programming algorithm, and outputting an optimal equipment configuration and construction scheme of each stage. According to the method, the problems of load increase unpredictability and energy market price fluctuation risk in different development stages of the park energy system are solved by combining scene analysis, data-driven modeling and a dynamic optimization mechanism.

The invention discloses an extreme high-temperature scene multi-energy complementary optimization scheduling method considering uncertainty, which mainly comprises the following steps of: 1, fusing bidirectional time convolution, a bidirectional long-short-term memory network, an attention mechanism and a quantile regression forest, realizing high-precision prediction and uncertainty modeling of wind speed, solar irradiation and load, and constructing a typical day scene set; 2, constructing a two-stage scheduling model fusing epsilon-constraint multi-objective optimization and opportunity constraint mixed integer programming, optimizing and adjusting margin day ahead, and rolling and correcting a scheduling path within the day; 3, three physical correction mechanisms of wind power air density correction, photovoltaic temperature response and hydroelectric evaporation-water level coupling are provided for extreme high-temperature disturbance; and 4, a prediction-optimization-feedback-correction closed-loop process is integrated and constructed, and the stability and response toughness of the system under extreme climate are improved. The method has the beneficial effect that the prediction precision, the scheduling flexibility and the operation toughness of the system under the extreme climate are remarkably improved.

The invention relates to the technical field of equipment life prediction, and provides an equipment life prediction method based on a mixed attention enhancement time sequence convolutional network, which comprises the following steps of: preprocessing original test data, extracting 10 types of time domain statistical characteristics from the preprocessed original test data, screening high-importance feature data as model input through a random forest algorithm; a life prediction model is constructed, an encoder adopts a stacked expansion causal convolutional layer and a self-attention layer, and a decoder fuses historical features and exogenous variables through cross attention; training a life prediction model by using a mixed attention enhancement time sequence convolutional network, wherein a composite loss function synchronously optimizes point prediction and multi-quantile regression loss; and inputting sensor data collected in real time into the trained model, outputting a prediction result, and generating a 95% confidence interval based on nonparametric probability prediction. The overall reliability and accuracy of equipment life prediction can be effectively improved.

The invention discloses a medical institution dynamic risk early warning method and device and a storage medium. Relates to the technical field of smart medical. According to the method, equipment parameters are collected in real time, and load self-adaptive dynamic threshold early warning is realized based on a quantile regression forest model; quantifying an influence score of the associated equipment by using a pre-constructed knowledge graph; and fault injection simulation is carried out by means of a digital twin model, the root cause is verified by comparing simulation data with real data, and finally accurate early warning information is pushed to a person in charge. According to the method, the problems of high false alarm and missing alarm rate and difficulty in multi-alarm root cause positioning of traditional fixed threshold early warning are solved, and accurate and automatic fault diagnosis and early warning are realized.

The invention relates to the technical field of network monitoring, in particular to a computer network data flow monitoring system and method, and the system comprises a data flow collection module, a protocol layer analysis module, a time sequence feature extraction module, a defect compensation reconstruction module and a monitoring result output module. According to the method, a continuous difference track can be formed by extracting protocol fields and marking change time points in data stream acquisition, a cross-layer synchronization index is generated by calculating the interval and the sequence of adjacent fields, so that a cross-layer relationship is quantized in a dynamic process, quantile regression is adopted in feature processing to unify a numerical interval, and the accuracy of the cross-layer relationship is improved. Compared with the prior art, the method guarantees the comparability under different time scales, can rapidly identify an abnormal section when cross-layer synchronization is broken, deduces a potential mode in combination with feature values before and after abnormity, reconstructs defect information in a probability mode, enables the result to have the abnormity recognition precision and the flow monitoring integrity, and improves the network management efficiency and safety.

The invention discloses an environmental benefit and resource load evaluation method for an urban sewage treatment plant, and belongs to the field of sustainable evaluation. The implementation method comprises the following steps: training a random forest model by taking pollutant effluent concentration, energy consumption intensity and medicament consumption intensity of a national town sewage treatment plant as output variables and taking other historical operation information as input variables; on the basis, quantile regression is used for calculating quantiles of actual values of three operation indexes of each factory under the original processing condition; constructing an environmental benefit evaluation index system represented by pollutant removal efficiency and a resource load evaluation system represented by consumption efficiency of resources such as energy and chemicals; through obtaining environmental benefit and resource load grading benchmark values of the town sewage treatment plant, core functions and necessary input of each plant are accurately quantified, and real environmental contribution of the town sewage treatment industry is clarified, so that evaluation of sustainability of the town sewage treatment plant is more objective and fair, and sustainable development of urban environment infrastructures is facilitated.

The invention discloses a risk indicator construction method and system for new energy output prediction, and belongs to the technical field of data processing and power system management, and the method comprises the steps: obtaining the preprocessing data of a plurality of new energy stations, carrying out the spatial-temporal feature analysis, and generating a spatial-temporal feature set; obtaining meteorological prediction data, and performing medium and long term output prediction through an extreme gradient boosting tree based on the space-time feature set; comparing the output prediction result with historical real output data to obtain a prediction error, and performing probability coupling modeling and quantile regression prediction in combination with a meteorological element evolution sequence to generate a dynamic quantile value; and determining a dynamic risk interval, and constructing a graded early warning index. According to the method, probabilistic coupling modeling and quantile regression prediction are adopted, prediction errors are deeply associated with dynamic evolution of meteorological elements, predicted uncertain risks can be quantized and graded, and decision support is provided for optimal scheduling and risk management of a power grid.

The invention discloses a photovoltaic power prediction method based on an adaptive modified quantile regression neural network. Feature extraction is realized by constructing a double-flow hybrid neural network so as to improve prediction accuracy, a branch, combined with a multi-head attention mechanism, of the convolutional neural network is responsible for extracting long-term features, a branch of a bidirectional gating circulation unit focuses on identifying short-term fluctuation, and the double-flow hybrid neural network is combined with quantile regression. In order to solve the problems of quantile crossing and non-differentiable zero point of a loss function, a self-adaptive correction marble loss function is provided, and smooth function optimization is introduced to ensure monotone increasing of predicted quantiles and whole-domain differentiable of the loss function. According to the method, point prediction, interval prediction and probability density prediction can be realized, the prediction effect is verified through a multi-dimensional evaluation index, potential information of photovoltaic power is fully mined, and the method has practical engineering application value.

The invention relates to the technical field of energy consumption interval prediction, and particularly discloses a building energy consumption interval prediction method, system and equipment, and the method comprises the steps: collecting historical energy consumption data, meteorological data, building characteristic data and use mode data of a target building, and carrying out the hierarchical processing, thereby obtaining a hierarchical data set; performing wavelet transform decomposition on the historical energy consumption data, and decomposing the energy consumption time sequence into a trend term, a periodic term and a random term to obtain multi-scale decomposition features; constructing a feature evaluation model, calculating a contribution degree and dynamically allocating weights to obtain a weighted feature set; constructing an integrated prediction framework based on the weighted feature set, and generating a prediction interval through quantile regression; and performing adaptive adjustment according to the historical prediction deviation and the environmental factor change to obtain a final prediction interval. Through multi-scale feature processing, dynamic weight optimization and prediction interval adaptive calibration, the accuracy and reliability of the prediction interval are improved, and more accurate support is provided for building energy management.

The invention provides an optical access network PON port fault prediction and operation and maintenance support method and system, and the method comprises the steps: collecting real-time multi-index sequences of optical power, bit error rate, temperature, frame loss rate, business load and the like, automatically extracting a long-term trend based on variational mode decomposition, constructing a residual sequence, generating a state transition map through employing a Markov transition field algorithm, and carrying out the fault prediction and operation and maintenance support of the PON port. And future operation state distribution prediction is realized through the LSTM neural network. And after an abnormal threshold is dynamically adjusted in combination with quantile regression, a time sequence prediction result and a threshold response are fused to output a comprehensive fault probability score, and an early warning signal is triggered through exponential smoothing and continuity check, so that the progressive fault detection accuracy and advancement can be improved, and an efficient and steady prediction and early warning basis is provided for intelligent operation and maintenance of the PON port.

The invention belongs to the field of geological disaster prediction, and particularly provides a reservoir landslide displacement prediction method based on dynamic lag recognition and fuzzy entropy optimization, which comprises the following steps: acquiring and preprocessing landslide time sequence monitoring data; combining the distributed lag nonlinear model with the maximum information coefficient, dynamically analyzing the lag relationship between the displacement and the rainfall and reservoir water level through a sliding window, outputting a self-adaptive lag stage and constructing a lag feature set; adaptively decomposing the displacement sequence by using variation mode decomposition of fuzzy entropy optimization, determining the optimal mode number according to the minimum fuzzy entropy, and reconstructing the intrinsic mode function into trend, period and random items; the method comprises the following steps: extracting local features of a multi-lag feature space through CNN, inputting reconstructed displacement components into GRU to capture time dependence, introducing an attention mechanism to weight a key time step, and outputting a predicted value and a confidence interval through quantile regression; according to the method, dynamic lag capture, adaptive decomposition and CNN-GRU-Attention are fused, and high-precision and high-robustness prediction is realized.

The invention provides a wind power prediction method based on marine meteorological characteristics and wake flow interference, and aims to solve the problem that the existing wind power prediction does not fully consider the thermal and dynamic characteristics of a marine boundary layer, the wake flow interference effect of a fan group and the attenuation of a salt mist environment. According to the method, power prediction is realized by constructing a multi-source marine meteorological characteristic field, a wake flow interference field and an environment attenuation compensation mechanism. Specifically, a meteorological characteristic matrix is generated by combining an integral norm of wave height change, a position and wind direction characteristic matrix is constructed based on a fan position and a prevailing wind direction, wake flow interference intensity is quantified through a Gaussian kernel function, and meanwhile, an attenuation compensation item containing an error function is designed to process a salt spray corrosion effect. Establishing a bimodal wind speed probability distribution model, and combining a space-time adaptive quantile regression output power prediction interval; the method effectively solves the problems of marine meteorological dynamic characterization, wake flow interference quantization and environmental attenuation compensation, and significantly improves the precision and reliability of offshore wind power prediction.

This invention relates to the field of power system operation and planning technology, specifically to a short-term wind power segment probability prediction method. It utilizes deep learning to mine implicit information in data and nonlinear features in wind power sequences to generate prediction probability intervals. Simultaneously, a nonlinear weighting method is selected to improve the optimization performance of the particle swarm optimization algorithm (IPSO), addressing some problems of traditional algorithms and improving convergence speed. Furthermore, a hybrid artificial intelligence algorithm, CNN-LSTM, is selected to construct a prediction model based on a combination of SVM and quantile regression (IPSO-CNN-LSTM). After training, short-term wind power probability prediction is completed. The CNN network can extract latent features from sample data using convolutional kernels, while the Long Short-Term Memory (LSTM) network can capture long-term components, avoiding gradient vanishing and exploding phenomena found in some existing algorithms, thus improving the efficiency of wind power probability prediction.

The application is a kind of biomarker detection data processing method and system, relating to data classification technical field, including acquiring multi-source biomarker data, using wavelet-quantile regression to denoise, constructing sample correlation graph, and filling in missing values by graph attention network.Deep forest combines causal graph to learn features, cascades multi-granularity scanning to capture nonlinear interaction, combines permutation importance and SHAP value to screen markers.Multi-modal encoder extracts high-dimensional features, multi-modal attention fusion network fuses features, cross-modal attention mechanism aligns, and generates joint representation vector.Multi-task contrast learning optimization introduces contrast loss function to obtain cross-modal fusion features.Input classification model prediction obtains biomarker classification results.

This application provides a method, system, and storage medium for quantitatively assessing a building's flexible energy utilization capacity. The method includes: collecting historical building operation data and constructing input features; based on the input features, establishing a building load behavior baseline prediction model using quantile regression to predict the typical load operation baseline and its upper and lower fluctuation amplitudes at each time point; constructing a similar sample set based on samples from the historical operation data sample that have similar operation attributes to the time point to be assessed; taking the high and low quantile values ​​of the load distribution from the similar sample set as the original upper and lower boundaries of the load at the time point to be assessed, and correcting the upper and lower fluctuation amplitudes of the typical load operation baseline at the time point to be assessed to obtain the final feasible upper and lower boundaries; calculating the instantaneous downward and upward adjustment power based on the load behavior baseline prediction model and the feasible upper and lower boundaries; calculating the continuous downward and upward adjustment energy within the control time window; and obtaining the building's flexible energy utilization capacity index.

The application discloses a spare part prediction method based on a PSO training quantile neural network, combines a quantile regression neural network (QRNN) with a recurrent neural network (RNN) to construct a spare part demand prediction model, can more comprehensively capture the conditional distribution characteristics of the spare part demand through quantile regression analysis, and further provides deep insight into demand fluctuation, meanwhile, the introduction of the RNN enables the model to effectively process the time sequence dependency in time sequence data, so that the influence of past demand on current demand is considered in the prediction process, the efficiency of spare part management is effectively improved, the inventory cost is reduced, the quality and satisfaction of customer service are improved, important economic benefits and competitive advantages are brought to the manufacturing and service industries, therefore, the application has wide application prospects and can play an important role in various electronic products and after-sales services.

The invention provides a new energy power system reliability evaluation method based on environment variables, and relates to the technical field of power systems. According to the method, a non-parametric self-adaptive quantile regression forest model is adopted to perform modeling on wind speed and sunlight intensity, and online incremental updating is triggered based on Kolmogorov-Smirnov statistics; setting a cross-scale state coupling interface layer in the double-layer nested reliability evaluation architecture, defining a shared state vector including an energy storage charge state, an equipment aging index and an available reserve capacity, and applying Lyapunov stability constraint; a period economic value dynamic pricing module based on Q-learning is integrated in a short-term evaluation layer, and a real-time electricity price, a user interruption contract and a load elastic coefficient are used as state spaces to correct a load shedding strategy. According to the scheme, through adaptive modeling of environment variable distribution and cross-scale state stable transmission, the problem of evaluation result lag is effectively solved, and the real-time performance and economical efficiency of reliability evaluation are improved.

This invention belongs to the field of job scheduling, and relates to a method, apparatus, computer equipment, and storage medium for scheduling jobs in a computing cluster. The method includes: acquiring multi-source job data; preprocessing and constructing multi-dimensional features from the multi-source job data; constructing a heterogeneous ensemble learning model and performing quantile regression training; based on the quantile regression training results, quantifying prediction uncertainty and generating job runtime estimates and probabilistic prediction intervals; transforming the job runtime estimates and probabilistic prediction intervals into a set of budget uncertainties and establishing a robust optimization scheduling model; transforming the robust constraints in the model into equivalent deterministic linear constraints; embedding the robust optimization scheduling model into a rolling time-domain control framework, and combining it with a prediction error feedback mechanism to achieve online adaptive scheduling. This achieves the proactive quantification and utilization of job runtime prediction uncertainty; improves the reliability and performance stability of the system under uncertain environments; and enhances reliability.