217 results about "Parameter learning" patented technology

The invention discloses a watershed water resource supply and demand risk prediction and evaluation method based on a multilevel Bayesian causal network, and relates to the technical field of water resource supply and demand risk management.The watershed water resource supply and demand risk prediction and evaluation method comprises the steps that a water resource supply and demand risk diagnosis knowledge graph is constructed according to key variables and interrelations input by a user; constructing a multi-level Bayesian causal network structure; estimating conditional probability distribution among the nodes, and performing parameter learning and structure training on the Bayesian causal network; carrying out risk path identification through a reverse Bayesian reasoning method; outputting a posterior probability of water resource supply and demand risk prediction; based on a preset fuzzy character string matching algorithm, typical risk events and risk features are extracted; and according to the posterior probability and the risk characteristics, comprehensively evaluating the water resource supply and demand risk level. The method can improve the systematicness and scientificity of risk identification, is suitable for multi-link and multi-scale risk assessment and scheme comparison and selection in a complex drainage basin, and has high practical value and popularization prospect.

The invention discloses a method and system for predicting juvenile depression based on intestinal flora, and relates to the technical field of bioinformatics and artificial intelligence, and the method comprises the steps: firstly, obtaining an original sequence of a microbiome, carrying out the preprocessing of the original sequence of the microbiome, and obtaining a feature matrix; and screening core flora characteristics with stable trans-folding by adopting characteristic importance evaluation and interpretability analysis based on a gradient boosting decision tree. A mixed weighted graph is constructed based on Spearman correlation and a proximity relationship, and an absolute value of a correlation coefficient is taken as an edge weight and an edge density is adjusted through a threshold adaptive strategy. And finally, through an improved graph attention neural network, based on edge weight attention, layer normalization and random inactivation, enhancing robustness, and adopting adaptive optimization to complete parameter learning. And determining a dynamic classification threshold according to the AUC of the target patient, and outputting a sample discrimination result and confidence. According to the method, the accuracy, stability and biological interpretability of juvenile depression recognition are remarkably improved.

The invention belongs to the technical field of intelligent education, and discloses a multi-modal artificial intelligence learning assistant based on an embedded platform, and the method comprises the steps: a data collection and processing module receives learning problems and student learning data, and carries out the standardization processing, and obtains a problem standardized data set; the cognitive model construction module constructs a learner cognitive model based on the problem standardized data set and student learning data, and marks knowledge gap nodes; the teaching scheme generation module determines a self-adaptive teaching scheme including knowledge depth parameters, a learning material combination strategy and an explanation expression strategy according to the structural features of the cognitive model; the learning sequence construction module constructs a customized learning sequence according to the knowledge gap nodes and the knowledge depth parameters; and the learning content generation module generates personalized learning content according to the learning sequence and the teaching scheme. Through the self-adaptive processing flow driven by the cognitive model, personalized teaching for different students is realized, and the learning efficiency and the learning effect are remarkably improved.

The invention belongs to the technical field of tunnel construction risk assessment, and provides a shield tunnel excavation face instability disaster risk quantitative assessment method and system. The evaluation method comprises the following steps: constructing a shield tunnel excavation face instability disaster risk evaluation system and converting the system into a Bayesian network topological structure; calculating the membership degree of each disaster evaluation index according to the digital feature calculation of the forward cloud generator and the to-be-evaluated data, and converting the membership degree into a Bayesian network prior probability; according to the historical data of the instability disaster of the excavation face of the shield tunnel and the Bayesian network topological structure, parameter learning is conducted on the historical data of the instability disaster of the excavation face of the shield tunnel, and the conditional probability of a Bayesian network model is calculated; according to the Bayesian network prior probability, the Bayesian network model conditional probability and the Bayesian network topological structure, calculating a quantitative evaluation result of the instability disaster risk of the excavation face of the shield tunnel; and according to the quantitative evaluation result of the instability disaster risk of the excavation face of the shield tunnel, performing adjustment decision on shield tunneling control parameters.

The invention provides a sewage treatment water quality parameter real-time detection system based on deep learning, and relates to the technical field of data processing, and the system comprises a data collection module which is used for collecting multi-source dynamic data in a sewage treatment process in real time through a distributed sensor network; the feature reconstruction module is used for performing feature space reconstruction on the multi-source dynamic data and generating dynamic correction parameters through time sequence correlation analysis; and the learning prediction module is used for inputting the dynamic correction parameters into a pre-trained multi-task deep learning model, analyzing and evaluating the influence degree of each feature variable through the contribution degree of the parameters, and dynamically adjusting feature importance distribution by adopting a self-adaptive weighting mechanism so as to obtain adjusted feature representation. According to the invention, real-time accurate detection of water quality parameters, timely early warning of standard exceeding risks, energy consumption optimization in a sewage treatment process and stable control of effluent quality are realized.

The invention relates to an Internet of Things time sequence root cause analysis method based on a dynamic causal diagram, and belongs to the technical field of Internet of Things. The method comprises the following steps: embedding a fine-tuning large language model by utilizing an Internet of Things knowledge graph, embedding, splicing and constructing a graph structure through entities and relationships, and combining text word embedding and mask prediction task optimization model; on the basis of knowledge graph subgraph construction, a triple is converted into a natural language to be input into a large model to generate a causal hypothesis; assumptions are converted into causal constraints, dynamic causal graph structure learning is carried out in combination with a Bayesian information criterion scoring function, and conditional probabilities of father nodes and historical values are modeled; performing parameter learning by adopting kernel density estimation, and quantifying time hysteresis among the features; feature probability distribution is predicted based on sliding window observation data, accumulative error contribution is calculated through asymmetric Shapley values, and causal ancestor features are preferentially sorted to determine root causes. Real-time and effective root cause analysis of the Internet of Things system is realized, and the system has the capability of intelligently solving faults.

The invention discloses a scanning radar super-resolution imaging method based on data-driven adaptive parameter learning, which is applied to the technical field of radar imaging and aims to overcome the defects that a traditional super-resolution method is high in parameter sensitivity and an existing network expansion method has a redundant structure and poor adaptability. The method comprises the following steps: firstly, converting a non-differentiable L1 regularization problem into a differentiable L2 problem based on a maximum-minimum principle; secondly, expanding an iteration process into a multi-stage cascade neural network, and performing adaptive learning on regularization parameters in a data driving mode; and finally, layered parameter optimization and super-resolution reconstruction are realized through an end-to-end network architecture. By adopting the method of the invention, the super-resolution imaging performance of the radar is effectively improved.

The invention discloses a vehicle fault diagnosis method, system and device based on a Bayesian network and a medium, and the method comprises the steps: constructing the Bayesian network according to preset vehicle data variables and fault root cause description, and initializing a conditional probability table of each node and the attention weight of each connection edge; inputting preset multi-source sensor sample data into the Bayesian network for parameter learning, and optimizing the conditional probability table of each node and the attention weight of each connection edge to obtain a vehicle fault diagnosis model; inputting the target multi-source sensor data into the vehicle fault diagnosis model to obtain target probability distribution of a plurality of target fault root causes, and determining the contribution degree of each input variable to each target fault root cause according to the attention weight; and generating a fault diagnosis thermodynamic diagram according to the target probability distribution and the contribution degree, and marking a key input variable and a corresponding causal chain in the fault diagnosis thermodynamic diagram. The method improves the intuition and interpretability of the diagnosis result, and can be applied to the technical field of fault diagnosis.

The invention discloses a water quality probability forecasting method based on Bayesian multi-time-sequence deep learning. The method comprises the following steps: S1, determining a forecasted water environment water ecological index, a driving index and a forecasting day number; s2, collecting time sequence data monitored by the forecasting indexes and the driving indexes, and after data preprocessing, constructing a data set required by model construction; s3, carrying out data division on the time sequence data, constructing a driving index forecasting model by adopting a multi-time sequence deep learning method, and carrying out parameter learning by selecting a Bayesian random discarding method; s4, performing effect evaluation on the accuracy and precision of the model, and adopting a hyper-parameter optimization method to improve the simulation forecast effect; s5, carrying out model training by adopting all data without segmenting the training set and the test set, carrying out water quality probability forecasting by utilizing the trained model, and outputting a forecasting mean value and a confidence interval; according to the method, the confidence interval is output while high-precision prediction is provided, and the scientificity and stability of prediction are improved.

The invention relates to the technical field of unmanned tracked vehicle trajectory tracking, in particular to a learning type unmanned tracked vehicle trajectory tracking prediction control method, which comprises the following steps of: establishing an interactive training virtual environment oriented to tracked vehicle trajectory tracking controller learning; receiving vehicle global expected trajectory point information output by a tracked vehicle trajectory planning module; acquiring the motion states of the crawler and the obstacle at the current moment; a controller expected parameter instruction and a crawler longitudinal acceleration and yaw angle acceleration expected instruction are determined; a tracked vehicle longitudinal lateral motion tracking controller is constructed, and a vehicle longitudinal lateral motion bottom layer control instruction is determined; motion states of the crawler and the obstacle at the next moment are acquired; data are collected while interaction between the crawler and the environment is carried out, and a data experience pool for controller parameter learning is constructed; and constructing a predictive controller parameter reinforcement learning algorithm. According to the invention, through an autonomous exploration learning technology, the trajectory tracking precision of the unmanned tracked vehicle in a dynamic complex environment is improved.

The invention discloses a blurred image NeRF modeling method, system and device based on a scattering light path model and a medium. The modeling method comprises the steps of blurred image initial pose calculation, camera pose interpolation, neural radiation field NeRF-based 3D scene modeling, scattering light parameter learning guided by an internal scattering light path model, blurred image synthesis prediction by using a scattering sensing volume rendering method, and parameter joint optimization based on blurred image luminosity loss. The system, the equipment and the medium are used for implementing the method. A light propagation phenomenon is represented as internal scattering of light at a medium or surface intersection point through an internal scattering light path model, a scattering light path direction and a sampling point distance are autonomously learned by means of a feedforward neural network, and image rendering is performed by integrating contributions of sampling points in the directions of a main light path and the scattering light path. The real imaging process of a blurred image in a complex illumination environment can be effectively simulated, the geometric ambiguity problem is effectively avoided, the convergence stability of the neural network is improved, and a finer reduction effect on geometric details of a fine object is achieved.

The invention discloses a fire safety assessment method and system based on machine learning, and relates to a machine learning and fire safety assessment technology, and the method comprises the steps: collecting image data, equipment operation data and environment monitoring data of a fire-fighting place; inputting the first image feature vector into a pre-trained deep convolutional neural network model to obtain a risk level label, and inputting the first standardized data into a long-short term memory network model to obtain a first time sequence prediction result; performing feature fusion on the first risk level label and the first time sequence prediction result, constructing a fusion feature matrix, and inputting the fusion feature matrix into a random forest model for parameter learning; and inputting newly collected fire-fighting data into the learned random forest model to carry out safety level prediction so as to obtain a fire-fighting safety prediction result and generate risk early warning information. According to the invention, risk early warning information can be generated in real time, fire-fighting potential safety hazards can be quickly responded, and the fire-fighting management efficiency is remarkably improved.

The invention provides an earthquake landslide disaster chain risk assessment method, medium and equipment, and relates to the technical field of geological disaster chain risk assessment, and the method comprises the steps: obtaining earthquake-induced disaster chain data, and constructing an earthquake landslide disaster chain knowledge graph; according to the knowledge graph, acquiring disaster influence factors, determining association among the disaster influence factors, and determining key disaster influence factors with relatively high association degree; key disaster influence factors are used as Bayesian network nodes, node variable data are discretized, a maximum-minimum hill climbing algorithm is used for structural learning, and a directed acyclic graph structure of the Bayesian network is determined; performing Bayesian network parameter learning by using an expectation maximization algorithm to obtain a prior probability of Bayesian network nodes; if the prediction result of the model does not reach the expectation, the association between the nodes is adjusted, and if the prediction result reaches the expectation, the model is used for earthquake landslide disaster chain risk assessment. According to the scheme, the reliability of the prediction result can be improved.

The present invention discloses a method for collaborative simulation of multiple water quality indicators based on a physical information deep neural network, comprising the following steps: S1, determining the water quality indicators to be collaboratively simulated, and constructing a multi-source database for multi-water quality indicator system simulation; S2, selecting a deep learning model suitable for multiple output water quality indicators, and completing the construction of a physical information deep neural network for multi-indicator water quality collaborative simulation; S3, using a k-fold cross-validation method to train the water quality collaborative simulation model, and completing parameter learning of the water quality collaborative simulation model; S4, selecting evaluation indicators to evaluate the water quality collaborative simulation model, and improving the simulation effect by adjusting the network structure and optimizing hyperparameters until the water quality collaborative simulation model meets the simulation accuracy requirements; S5, based on the optimized water quality collaborative simulation model, using a deep learning model interpretation method to analyze the key driving factors of the collaborative changes of multiple water quality indicators, and complete the multi-indicator water quality collaborative simulation.

The invention provides a loudspeaker mask defect identification method and system based on visual inspection, and relates to the technical field of machine visual inspection, and the method comprises the steps: collecting a surface image of a loudspeaker mask through an industrial camera, calculating a median gray value in a pixel neighborhood to effectively filter out impulse noise, carrying out size standardization processing, and obtaining an image of the surface of the loudspeaker mask; a standardized image is obtained; according to the standardized image, a convolutional neural network is adopted, and a parameter learning rate is dynamically adjusted through an adaptive moment estimation mechanism, so that multi-scale feature mapping of mask textures is extracted; according to the multi-scale feature map, candidate defect regions are generated through a region suggestion network; and for a candidate defect area proposal, using a U-Net network and a learning rate to quickly approach an optimal solution, attenuating the learning rate according to an exponential law along with the increase of the number of iterations to stably converge to fine local optimum, and performing pixel-level fine segmentation on a defect boundary to obtain a defect mask image. According to the invention, the detection efficiency of the loudspeaker mask is improved.

The invention provides a three-dimensional target detection method based on monocular vision, and belongs to the field of three-dimensional target detection for automatic driving, and the method comprises the following steps: designing an improved backbone network Faster Net +, carrying out image feature extraction, constructing a multi-dimensional feature adaptive fusion module, and adaptively selecting and fusing high-dimensional and low-dimensional features; a feature enhancement attention module is introduced on a multi-scale feature layer extracted by the feature pyramid network, interaction between feature channels and correlation between different coordinates are considered at the same time, a target area is highlighted, and irrelevant background information is inhibited; introducing an effective target detection head network according to the three-dimensional target detection network model, and performing parameter learning on the network model by using a training data set; after training is finished, a test image is input, and the positions and categories of different types of targets in the image are determined by using the three-dimensional target detection network model.

The invention discloses a multi-water-quality-index collaborative simulation method based on a physical information deep neural network, and the method comprises the following steps: S1, determining water quality indexes needing collaborative simulation, and constructing a multi-source database for multi-water-quality-index system simulation; s2, selecting a deep learning model suitable for multi-output water quality indexes, and completing the construction of a physical information deep neural network for multi-index water quality collaborative simulation; s3, training the water quality collaborative simulation model by adopting a k-fold cross validation method, and completing parameter learning of the water quality collaborative simulation model; s4, selecting an evaluation index to evaluate the water quality collaborative simulation model, and improving a simulation effect by adjusting a network structure and optimizing hyper-parameters until the water quality collaborative simulation model meets a simulation accuracy requirement; and S5, based on the optimized water quality collaborative simulation model, analyzing key driving factors of collaborative change of multiple water quality indexes by adopting a deep learning model interpretation method, and completing multi-index water quality collaborative simulation.

The invention discloses a light-weight cabin efficient self-adaption machining method based on cutting parameter learning, solves the cutting parameter self-adaption problem in light-weight cabin machining, and belongs to the technical field of spacecraft structural part manufacturing or numerical control self-adaption machining. The method aims at improving the machining quality and the machining efficiency, minimizing the cutting force and maximizing the material removal rate, a mathematical model of two targets and cutting parameters is established, an orthogonal process test is designed to analyze the response characteristics of the two targets to the cutting parameters, the influence of the cutting parameters to the cutting force in the actual cutting process is analyzed, and the machining precision is improved. Optimized operation data and a tool database are formed; and learning of adaptive parameters is realized by applying an ACM adaptive optimization technology, and adaptive processing is carried out.

An apparatus of a vehicle comprises a memory storing at least one instruction and a processor configured to execute the at least one instruction. The at least one instruction may be configured to cause, when executed by the processor, the apparatus to: via a tuning parameter learning model for image correction, generate, based on received video data, a tuning parameter for adjusting image signal processing (ISP) for correcting the received video data; correct, based on the tuning parameter, the received video data; identify, via a video recognition model, at least one object in at least one image corresponding to the corrected video data; and control, based on the identified at least one object, autonomous driving of the vehicle.

The invention relates to the technical field of power grid reactive voltage operation control, in particular to a power grid reactive voltage strategy regulation effect tracing reasoning method, device and equipment and a medium, and the method comprises the steps: constructing a regulation effect quantitative evaluation index system from safety, economy and schedulability; comprehensive weighting with subjective and objective combination is adopted, and key index weights are corrected; comprehensively evaluating each level of indexes of the typical day by using an ideal approach solution to form an operation state result; extracting an abnormal index based on a result, and constructing an associated factor set in combination with a regulation and control strategy and operation characteristics; screening strong correlation factors through correlation and causality analysis; k mean discretization is carried out on the time sequence of the abnormal indexes and the time sequence of the abnormal indexes, and Bayesian network parameter learning is completed by adopting maximum likelihood estimation; key causes are identified based on network reverse reasoning, and the method can accurately evaluate the regulation and control effect, locate the effect key factors and provide decision support for power grid dispatching optimization.

The invention relates to a pollution discharge enterprise equipment operation state identification method, device and equipment, and the method comprises the steps: generating a real-time power utilization data sequence through collecting the electrical parameters of power utilization equipment of a pollution discharge enterprise in real time; performing statistical analysis on the real-time power consumption data sequence in the sliding window, and calculating the power consumption change rate, the power factor and the start-stop frequency in the corresponding window; constructing an equipment operation state recognition model by using a dynamic Bayesian network, and independently constructing a sub-model for each process stage to respectively perform parameter learning; inputting the multi-dimensional continuous features into a dynamic Bayesian network, and outputting the start-stop state of the equipment at the current moment; and dynamically adjusting a threshold value for judging the start-stop state of the equipment. The dynamic change of the start-stop state of the equipment is flexibly handled, the problem that a fixed threshold cannot adapt to the operation fluctuation of the equipment is avoided, and the judgment precision of the start-stop state of the equipment is ensured; the difficulty of cluster number selection is avoided, and the diversity and complexity of the operation state of the equipment can be better reflected.

The invention relates to a cable tunnel bridge fire prediction method and device, electronic equipment and a storage medium. The cable tunnel bridge fire hazard prediction method comprises the following steps: establishing a cable tunnel bridge fire hazard numerical simulation model, establishing a cable tunnel bridge fire hazard simulation model based on FDS software, obtaining key parameters such as heat release rate, temperature distribution and flame spread boundary under different working conditions, and constructing a multi-time sequence sample data set; performing state discretization on continuous variables such as temperature and a spreading range, and estimating a node prior probability in combination with a statistical frequency; constructing a dynamic Bayesian network topological structure according to a causal relationship among a heat source, temperature and spread, and setting a cross-time slice node dependence path to realize time sequence modeling; inputting a training sample and performing parameter learning to generate a conditional probability table; in the prediction stage, multi-step reasoning is achieved through forward propagation, flame spreading range probability distribution and interval estimation of multiple time steps in the future are obtained, and the specific position where the flame arrives is predicted.

The invention discloses a radar interference effect evaluation method based on a constraint learning dynamic Bayesian network, is applied to the field of radar interference evaluation, and aims at solving the problem that the accuracy of interference effect evaluation is reduced due to radar detection data missing in a complex electromagnetic environment. Meanwhile, parameter constraints of five types of evaluation indexes and interference effect grades are defined; secondly, constructing a constraint learning dynamic Bayesian network, and learning a conditional probability and a transition probability under a data missing condition; then, proposing a prior constraint expectation maximization algorithm, converting parameter learning into an optimization problem with constraint by combining convex optimization, and overcoming the defects of a traditional expectation maximization algorithm; secondly, a cloud model is introduced to quantify discrete probability distribution into a continuous interference degree value; finally, simulation shows that the method can effectively improve parameter learning stability and evaluation accuracy under the conditions of suppressing and deception jamming and single index deficiency, and provides a reliable scheme for radar jamming effect evaluation in a complex environment.

The invention provides a learning rate adaptive bearingless magnetic flux switching motor neural network PID suspension method, which is used for PID magnetic suspension control of a motor PID controller on a rotor, and adjusts a neural network weight coefficient in real time according to a rotor radial displacement control error so as to realize real-time adjustment of parameters of the neural network PID controller. Establishing a neural network PID parameter learning rate value range according to a neural network PID closed-loop control stability requirement; within a learning rate value range, designing a self-adaptive learning rate adjustment algorithm based on fuzzy reasoning for high-steady-state control precision and high-dynamic response of the dynamic eccentric magnetic suspension of the rotor in a wide range; the method can meet the requirements of magnetic suspension high-steady-state control precision and high dynamic response of wide rotor dynamic eccentricity.

The invention discloses a Raman spectrum noise reduction method based on an MSC-U-Net algorithm, and the method comprises the steps: S1, measuring Raman spectrums of a plurality of substances with different concentrations under the same condition, and respectively measuring a plurality of spectrums for each concentration of each substance, so as to obtain an original spectrum data set containing noise; s2, performing normalization preprocessing on all the collected spectrums, dividing a data set into a training set, a verification set and a test set, and constructing a data set required by a noise reduction model; s3, averaging the collected original spectrums according to a plurality of spectrums of the same substance and the same concentration to obtain ideal spectrums without noise; and S4, constructing a Transform-Attention U-Net model, inputting the training set into the model to carry out parameter learning, and carrying out iterative optimization according to a loss function to obtain a spectral noise reduction model. According to the method, noise signals in the Raman spectrum can be effectively removed, meanwhile, effective signals are not lost, and the signal-to-noise ratio of the Raman spectrum is increased.

The invention provides a radar signal small sample modulation identification method and system based on meta-learning, and relates to the technical field of radar signal modulation identification, and the method comprises the steps: collecting a multi-polarization radar echo signal of a target object, and carrying out the preprocessing of the multi-polarization radar echo signal to generate a multi-polarization feature sequence and a modulation feature set; analyzing signal channel quality, performing multi-channel fusion, and obtaining target enhanced polarization characteristics through airspace interference suppression; performing multi-modal feature alignment on the modulation feature set and the modulation feature set, generating multi-modal feature representation through joint mapping and similarity measurement, and reducing intra-class difference; and finally, on the basis of a meta-learning framework, constructing a layered optimization framework to carry out parameter learning and meta-parameter adjustment, realizing adaptation and generalization of a radar signal modulation identification rule under a small sample by means of normalized loss mechanism standard training, and carrying out multi-polarization signal processing, multi-modal feature fusion and meta-learning optimization. Effective adaptation and generalization of the radar signal modulation recognition rule are realized under the small sample condition, and the recognition performance is improved.

The invention discloses a vehicle crawling control method, device and equipment and a storage medium, and the method comprises the steps: firstly detecting a driving mode selected by a driver, and then carrying out the matching from a pre-established crawling moment control logic set according to the driving mode, and executing a corresponding target control logic. On the basis, the system continuously collects operation parameters of the driver in the mode, and dynamically updates and optimizes the currently executed target crawling moment control logic by using the parameters, so that the self-adaptive adjustment of the control logic is realized. According to the method, by introducing a driving mode recognition and parameter learning mechanism, control over the crawling moment can adapt to different driving styles and scene requirements, and the economical efficiency, the NVH performance and the driving responsiveness of the whole vehicle are effectively improved. The method can be widely applied to the technical field of vehicles.

The invention provides a rocket aircraft flow field correction method based on multi-fidelity data fusion, and the method comprises the following steps: 1, collecting low-fidelity data, and carrying out the construction of a low-fidelity data set: simulating the change process of an unsteady flow field around an object in a finite time step through a Reynolds time-average simulation method or a large vortex simulation method; 2, data correlation analysis is carried out, wherein a linear relation model yH = rho (x) yL + delta (x) of a non-viscous flow field yL and a viscous flow field yH of the rocket aircraft is established; step 3, neural network architecture design: constructing a low-fidelity data approximation network NNL; 4, performing hyper-parameter learning and optimization; defining a loss function; 5, performing data acquisition and preprocessing: performing high-fidelity viscous flow field data acquisition, and preprocessing the acquired high-fidelity and low-fidelity flow field data; and step 6, model training and verification.

The invention discloses a quota management optimization method and system for medium and low voltage power protection projects in a distribution network, and relates to the technical field of quota management of medium and low voltage distribution networks, and the method comprises the following steps: calculating an initial quota base price of a target project sample according to a reference quota; based on the initial quota base price, the target project scene feature vector and historical project data, generating a corrected quota price through scene dynamic parameter learning; based on the target project scene feature vector and historical project data, generating a deviation estimation value through a scene deviation estimation model; determining a deviation compensation quota price according to the corrected quota price and the deviation estimation value; and taking the initial quota base price as a reference, carrying out weighted fusion on the increment between the enterprise quota price and the deviation compensation quota price according to confidence, and generating a final optimized quota price. The method is used for solving the problems that the difference between the quota price and the actual settlement price is too large in the quota management of the low-voltage power protection project, and the adaptability of the traditional quota to the complex construction scene is insufficient.

To provide a learning program, a learning method and an information processing unit that can learn a complicated probability distribution.SOLUTION: A computer is caused to execute: first processing to generate a probability distribution model by learning a probability distribution having less peaks than a target probability distribution; and second processing to generate a new probability distribution model by learning a probability distribution closer to the target probability distribution than the probability distribution having been learnt by the use of parameters of the probability distribution model having been generated.SELECTED DRAWING: Figure 4