63 results about "Correlation factor" patented technology

The invention discloses an integrated energy system short-term multivariate load prediction method fusing a deep learning combination model, and the method comprises the steps: obtaining the multivariate load historical data and meteorological data of an integrated energy system, and obtaining a correlation factor which is greatly affected by the multivariate load; dividing the multi-element load historical data into two parts, namely a training set and a test set, and performing maximum and minimum normalization processing; extracting multi-level spatial features of the multi-element load historical data by using a series-parallel superposed multi-scale convolutional neural network; load historical data time sequence features are extracted by using a bidirectional gating cycle unit neural network, and a multi-head attention mechanism is introduced to extract important information of a load prediction task, so that a multi-scale convolution bidirectional gating cycle unit neural network fused with the attention mechanism is designed; and a self-adaptive multi-strategy grey wolf optimization algorithm is utilized to realize optimization parameter search of the bidirectional gating circulation unit neural network, an evaluation index is adopted to carry out error analysis on a prediction result, an optimal parameter is obtained, and accurate load prediction is realized.

The invention discloses an intelligent maintenance system and method for an electric actuating mechanism based on fault prediction, and particularly relates to the technical field of equipment maintenance. The method comprises the following steps: acquiring historical operation data in a preset operation period, generating a historical health data set, extracting characteristic parameters to obtain a historical characteristic parameter set, and constructing a trend change characteristic matrix according to the change trend of the historical characteristic parameters; constructing a multi-dimensional state space by using the trend change feature matrix, calculating the state aggregation degree of the fault event in the space, and determining an early warning index set; meanwhile, collecting real-time operation data to extract real-time feature parameters, forming real-time feature state vectors, and performing correlation calculation on the real-time feature state vectors and early warning indexes in a space to obtain real-time risk correlation factors; and finally, generating a risk probability prediction model according to the real-time risk association factor, the historical characteristic parameter set and the trend change characteristic matrix, predicting a fault probability and outputting a maintenance scheduling suggestion, thereby realizing intelligent maintenance of the electric actuating mechanism.

The invention relates to the technical field of liquefied natural gas sampling, and discloses an automatic liquefied natural gas sampling method convenient to operate. The method comprises the following steps: acquiring historical sampling data in a preset sampling period and generating a data set; extracting historical sample characteristic parameters, and establishing a trend change characteristic matrix; constructing a multi-dimensional state space according to the matrix, and calculating the state aggregation degree of historical sampling abnormal events to determine a sample feature early warning index set; extracting real-time sample characteristic parameters and establishing vectors, and performing spatial position correlation calculation with the early warning index set to determine real-time risk correlation factors; and generating a risk probability prediction model in combination with the real-time risk association factor, the historical sample feature parameter set and the trend change feature matrix, predicting a sampling anomaly probability and generating an adjustment suggestion. According to the method, historical data and real-time analysis are utilized, the sampling accuracy and adaptability are improved, manual intervention is reduced, and a more reliable mode is provided for liquefied natural gas sampling.

The invention provides a method for exploring performance goodness and generalization ability of various machine learning models in runoff sediment simulation. According to the method, the time sequence change trend of a river long-time sequence hydrological factor, a meteorological factor, a vegetation factor, an artificial activity factor and a runoff sediment transport change correlation factor is determined. Then identifying the correlation between each factor and the runoff sediment change, and selecting an important influence factor as an input factor of a machine learning model; the performance goodness of different machine learning models in simulation of runoff and sediment transport is evaluated, a sample data set is randomly divided to eliminate the influence of limited sample size and sample sequence on a simulation result, and each model is repeatedly trained and tested to determine the empirical risk, predictive ability and generalization ability of each model. And sorting the contribution degrees of the input factors in the runoff and sediment transport simulation process by applying a local interpretation method so as to reveal the main driving factors influencing the water and sediment change of the drainage basin.

The invention discloses a port disaster risk grading method, which relates to the technical field of risk grading, and further considers comprehensive correlation coefficients among all consequences in each disaster type in a data-driven manner on the basis of the prior art, so as to ensure that correlation factors among different consequences can be comprehensively considered, and improve the risk grading efficiency. Meanwhile, a first composite index and a second composite index of each disaster type are analyzed and calculated, and a third composite index is obtained on the basis of combining the first composite index and the second composite index so as to ensure that complexity factors of consequences of each disaster type are effectively considered, and finally the obtained result is incorporated into an existing calculation formula of an expected loss value, so that the calculation accuracy of the expected loss value is improved. According to the method and the device, the probability of occurrence of risk level misjudgment can be effectively reduced to the greatest extent even if the risk level of disaster types with the same coordinates is judged subsequently, so that the expected loss value which is more accurate, more comprehensive and more scientific is ensured to be obtained, and the probability of occurrence of risk level misjudgment can be effectively reduced to the greatest extent.

The invention relates to the technical field of data processing, and discloses a method, device and equipment for data processing, a medium and a program product, the method comprises the following steps: in response to an analysis element determined by a user, obtaining analysis data corresponding to the analysis element and an associated factor corresponding to the analysis element; on the basis of the analysis data, contribution weights of the correlation factors to the target indexes are generated, an initial analysis result is obtained, and on the basis of a data correlation model, a mapping relation between the analysis elements and the target indexes is established; in response to the adjustment of the contribution weight, generating an updated analysis result based on the adjusted contribution weight and the analysis data; or in response to adjustment of the analysis elements by the user, second analysis data corresponding to the adjusted analysis elements are obtained, a target analysis result is generated based on the second analysis data and the contribution weight, and the analysis elements and the adjusted analysis elements have the same associated factors. Analysis element adjustment is supported, and the method can adapt to actual service scene change.

Disclosed is a method for generating control data for automated, computerized control of following vessel (204, 306) in convoy of waterborne vessels sailing in body of water using an artificial intelligence model, method comprising: providing first digital twin (DT) of lead vessel (202, 302) and second DT of following vessel, providing third DT of body of water, providing mission specific information for lead vessel, simulating mission using mission specific information, Al model, and one of following: first DT (304) of lead vessel, second DT of following vessel, third DT of body of water, executing first part of mission, generating control data based on simulation and execution of first part of mission, calculating correlation factor between simulation and execution of first part of mission, and when correlation factor is higher than predetermined value, then using generated control data and second DT of following vessel for controlling following vessel.

The invention discloses a distribution network material demand multiple regression prediction technology based on correlation factor analysis. Under the background of rapid development of economy and continuous acceleration of urbanization pace, an electric power distribution network project serves as a core part of infrastructure construction, and the importance of demand prediction of the electric power distribution network project is increasingly prominent. The demand of power distribution network materials is influenced by various factors, such as economic growth, population change, industrial structure, technical progress and the like, and the related historical data accumulation sample size is small. If only the historical time sequence data of the material quantity is predicted, on one hand, the prediction effect is limited by the existing small data sample size, and on the other hand, the prediction result is not accurate enough due to neglecting the influence of other important influence factors. According to the method, four related factors of the distribution network are comprehensively considered based on correlation analysis, the four related factors comprise the number of projects, the issued plan amount, the power supply amount and the GDP, all the influence factors are combined, and the distribution network materials are predicted through multiple regression. According to the invention, auxiliary support can be carried out on material purchasing related work of a material department, the purchasing cost and the inventory cost are saved to a great extent, and the enterprise competitiveness is improved.

The invention discloses a document collaborative analysis method and system based on deep learning, and relates to the technical field of deep learning. The method comprises the following steps: extracting quantitative characteristics from literatures, screening and constructing a literature internal analysis event set, and determining an external interference event set at the same time; the method comprises the following steps: determining initial states of two types of events by adopting a single index method, constructing an event influence matrix, and determining a cross influence coefficient according to a mapping rule by calculating a co-occurrence correlation factor, a domain specificity factor and a trend momentum factor; and on the basis of the event influence matrix, meeting normalization constraint, calculating a time development index to quantify the event comprehensive influence, and updating the event state to predict the development change trend of the analysis event in the literature in the time domain. And fitting the initial state of the event with the predicted state value, and drawing a development trend chart to assist literature research. According to the method, internal and external factors are comprehensively considered, the complex relation of the literature is mined, the literature development trend is accurately predicted, and the defects of a traditional method are overcome.

The invention relates to the technical field of charging pile operation and maintenance, and discloses a charging pile intelligent early warning operation and maintenance method and system based on big data. The method comprises the following steps: acquiring multi-dimensional operation data of the charging pile, and integrating to generate an operation data warehouse; extracting features from the warehouse to obtain a feature parameter set, and building a trend change feature matrix according to the time change trend of the parameter set; and constructing a multi-dimensional state space according to the matrix, and calculating the state aggregation degree of historical fault events in the space to determine a feature early warning index set. Acquiring real-time operation data, extracting parameters, establishing a real-time feature state vector, and calculating the spatial position correlation between the real-time feature state vector and the early warning index set to obtain a real-time risk correlation factor; and generating a risk probability prediction model in combination with the association factor and the trend matrix, predicting a fault probability, and judging whether to early warn and generate a maintenance scheduling suggestion according to the fault probability. According to the method, real-time monitoring, accurate early warning and efficient operation and maintenance of the charging pile are realized through big data, and reliable operation of equipment and charging experience of a user are guaranteed.

The present invention provides a method for perceiving and identifying icing risks of ultra-high-speed transmission channel lines, which belongs to the technical field of risk monitoring. The method specifically comprises: based on the line loss data of a reference date, determining the deviation of the line loss data in each transmission power interval on different reference dates, screening the specified dates in the reference date based on the deviation, determining the change of the weather data on the specified date, and combining the change of the line loss data in each transmission power interval on the specified date to determine the change correlation factor between the line loss data and the weather data on different specified dates; when determining the existence of an associated date using the change correlation factor, according to the matching results of the weather data of different associated dates with the current date, when determining that there is an icing risk in an abnormal line loss area, perceiving the icing risk based on the weather data in the abnormal line loss area, thereby improving the efficiency of line loss identification processing.

The invention discloses a material cladding layer performance prediction method and system based on big data. The method comprises the steps of data acquisition, data preprocessing, feature information processing, construction of a material cladding layer performance prediction model and material cladding layer performance prediction. The invention belongs to the technical field of data processing. According to the scheme, a transaction value is generated, a correlation factor, a primary increment and an advanced increment are calculated, a dynamic fluctuation factor is designed, an individual position is updated, and an optimal feature subset is determined according to a fitness value threshold value and the maximum search frequency; generating nodes, setting initial weights and embedded vectors, calculating weighted Euclidean distances and average weighted similarities, setting update state labels and update state factors, updating important thresholds, screening out important neighbor nodes, updating the embedded vectors and weights of the nodes, training weak classifiers, and performing weighted combination, so as to obtain a weak classifier; the feature subset most beneficial to material cladding layer performance prediction is found, and the accuracy and reliability of material cladding layer performance prediction are improved.

This invention discloses a USV berthing and unberthing pose estimation method based on ship-shore LiDAR collaboration, comprising: acquiring shipborne and shore-based point clouds and shore-based fixed reference poses; performing odometer calculations on continuous shipborne point clouds to generate LiDAR odometer factors; constructing a global descriptor for the ship-shore point clouds and performing scene matching to generate candidate ship-shore point cloud pairs; extracting key points and calculating FPFH descriptors, solving for high-precision ship-shore relative poses through coarse / fine registration, obtaining the final ship-shore relative poses through asynchronous reconstruction and generating correlation factors; constructing a collaborative factor graph model containing odometer and ship-shore correlation factors; solving the model and minimizing the overall residuals to obtain a globally consistent optimal pose, thereby achieving berthing and unberthing control. This invention improves ship-shore data transmission efficiency and matching accuracy through global descriptor matching and FPFH feature compression; and eliminates the cumulative error of single-ship odometers by leveraging shore-based global pose constraints, achieving high-precision continuous pose estimation under long-term navigation and ensuring the robustness of berthing and unberthing navigation.

The present invention discloses a method, system, device and storage medium for abductive reasoning of abnormal indicators of power grid dispatch post-evaluation, and the method comprises: obtaining time series data during power grid operation, extracting abnormal indicators and various related factors according to the time series data; calculating the correlation between abnormal indicators and various related factors, and obtaining strong related factors; calculating the causal relationship between abnormal indicators and various related factors, and obtaining strong causal factors; constructing a Bayesian network based on strong related factors, strong causal factors and abnormal indicators; performing parameter learning on the Bayesian network, and performing abductive reasoning of abnormal indicators based on the Bayesian network after parameter learning. The present invention can solve the key technical problem of abnormal indicator dominant factor analysis.

The invention discloses a wind power generator stator winding fault diagnosis method, and relates to the technical field of wind power generators. The method comprises the following steps: detecting winding end surface roughness, environment humidity and vibration data features which are combined into a first risk-induced feature set, and detecting surface roughness and vibration data features which are combined into a second risk-induced feature set at the joint of a straight line segment and the end; extracting a first correlation factor between historical risk correlation data and a historical coverage area value in the historical risk detection data, obtaining a second correlation factor to which the connection part of the straight line segment and the end part belongs according to the first correlation factor, and obtaining a first risk induction feature set according to the first correlation factor and the first risk induction feature set; and estimating an insulativity reduction risk index and a structural fatigue aggravation risk index to which the end part of the winding belongs currently, and estimating a risk temperature value of the connection part of the straight line segment and the end part after the temperature rise is generated currently according to the second correlation factor and the second risk-induced feature set. The risk diagnosis efficiency can be improved.

The invention discloses a resident energy-saving effect evolution prediction method and system, and the method comprises the steps: obtaining user preferences, user energy-saving knowledge levels and user education levels in a resident community, obtaining the number of household electric appliances of a user, and obtaining the data of a power transmission line in the resident community; on the basis of an entropy weight method, entropy values of user preferences, user energy-saving knowledge levels, user education levels and user household electric appliance numbers are calculated, and user load data are obtained through calculation according to the entropy values; performing transformation processing on the user load data and the power transmission line data based on Fourier transform, and analyzing correlation characteristics between the user load data and the power transmission line data to obtain a load correlation factor and a power transmission line correlation factor; and performing association prediction analysis on the user load data based on the load association factor and the power transmission line association factor to obtain load characteristic data. According to the resident energy-saving effect evolution prediction method provided by the embodiment of the invention, accurate detection of load prediction is realized.

The present application relates to a kind of energy efficiency dynamic monitoring and evaluation technical field, propose a kind of ship overall technical energy efficiency dynamic monitoring and evaluation method and system, can eliminate environmental influence, realize the transparent, fair, dynamic evaluation of ship " pure technology " energy efficiency state. Including the following steps: S100: carry out quasi-steady state trend test: for the parameter susceptible to ship maneuvering, system uses sliding time window to combine linear regression slope test method to distinguish;S200: carry out environmental self-adaptive threshold adjustment: for the fluctuation screening of speed, shaft speed and other parameters, system introduces the correlation factor of sea state, no longer uses fixed standard deviation threshold;S300: carry out physical validity final screening: the data reserved after above-mentioned step, finally carries out physical boundary check;S400: through core calculation engine module, input cleaned data set, calculate output index and its confidence interval;S500: carry out fault root cause analysis;S600: complete evaluation.

The present application relates to the technical field of smart grid, disclose a kind of multi-factor associated return home degree winter power load pre-control method and system, method includes: the historical power consumption data obtained in advance is transformed to obtain power load data, and power load data is identified to obtain power consumption problem;The influencing factor is obtained by analyzing power consumption problem, and the strong correlation factor data matrix is obtained by analyzing influencing factor, and the correlation influence coefficient is obtained by analyzing strong correlation factor data matrix;The trained power load prediction model is obtained by iterative training to power load prediction model, and the current power consumption data obtained in advance is input into trained power load prediction model to obtain load prediction data;The red yellow green classification area is obtained by dividing power grid;Power load pre-control scheme is formulated, so as to carry out power consumption prevention and control management.The present application can solve the problem that power grid overload is difficult to prevent and control during return home degree winter and power consumption prediction is not accurate enough.

The embodiments of the present invention provide a monitoring method, device, electronic device and computer-readable storage medium, which relate to the field of remote sensing technology. By obtaining the heat island effect image map and temperature data in the meteorological data, and then determining the target area in the heat island effect image map according to the monitoring range, the average temperature value of the target area is greater than the average temperature value within the monitoring range. Further, a target correlation factor set corresponding to the target area is obtained, and the target correlation factor set includes multiple target correlation factors. The target correlation factor represents the correlation factor that meets the strong correlation condition of the heat island effect. Finally, the heat island effect contribution value of the target correlation factor is calculated, and the heat island effect contribution value represents the influence of the corresponding target correlation factor on the heat island effect of the target area, thereby realizing a multi-angle analysis of the causes of the urban heat island effect.

The invention relates to the technical field of transformer anomaly detection, and discloses a transformer internal insulation state anomaly detection method, device and equipment and a medium, and the method comprises the steps: calculating a correlation factor set between data sequences collected by measurement points at any two different positions in a transformer; establishing an input vector; obtaining a predicted value of the data collected by each measuring point in the next day through an informer model; constructing a prediction sequence based on the prediction values of the continuous preset days and the association factor set, and constructing a real data sequence based on the real data actually collected by each measurement point of the continuous preset days and the association factor set; the difference value between the prediction sequence and the real data sequence is calculated, whether the internal insulation state of the transformer is abnormal or not is judged based on the difference value, and the accuracy of internal insulation state detection of the transformer is improved.

The application relates to the technical field of charging pile operation and maintenance, and discloses a charging pile intelligent early warning operation and maintenance method and system based on big data. The method comprises the following steps: acquiring multi-dimensional operation data of a charging pile, integrating and generating an operation data warehouse; extracting a feature parameter set from the warehouse, and building a trend change feature matrix according to the time change trend of the parameter set; constructing a multi-dimensional state space according to the matrix, calculating the state aggregation degree of a historical fault event in the space to determine a feature early warning index set. Real-time operation data is acquired, parameters are extracted to establish a real-time feature state vector, the spatial position correlation of the real-time feature state vector and the early warning index set is calculated to obtain a real-time risk correlation factor; a risk probability prediction model is generated by combining the correlation factor and the trend matrix, the fault probability is predicted, and it is judged whether early warning is needed and a maintenance scheduling suggestion is generated according to the fault probability. The method realizes real-time monitoring, accurate early warning and efficient operation and maintenance of the charging pile by means of big data, and guarantees reliable operation of the equipment and charging experience of the user.

The present invention discloses a method and a storage medium for carbon emission accounting and multi-dimensional analogy during the operation period of a sewage treatment plant, which relates to the technical fields of environmental protection governance and clean energy resources. The method includes: a program of selecting a statistical period, collecting activity information, checking and correcting data, selecting correlation factors, starting an accounting program, and showing index results, so as to obtain index results; when the index results establish a longitudinal evaluation object with a single sewage treatment plant, conducting carbon efficiency analysis and evaluation, weight influence ranking, and cross-analogy analysis of single-plant sources to obtain the evaluation results of the single sewage treatment plant; when the index results establish a horizontal evaluation object with multiple sewage treatment plants, conducting correlation trend research and judgment of multi-plant sources and cross-analogy analysis of multi-plant sources to obtain the evaluation results of multiple sewage treatment plants; and making decision optimization according to the evaluation results of the single sewage treatment plant and the evaluation results of multiple sewage treatment plants. The present invention conducts comparative analysis and low-carbon decision-making evaluation on the carbon emission indicator indexes of different periods or different sewage treatment plants of the sewage treatment plant.

The application provides a candidate item priority determination method for a civil aviation system, and relates to the technical field of data processing. The method first acquires a basic push score matrix according to a historical data set, then modifies the basic push score matrix in combination with a score adjustment weight matrix corresponding to a target user to realize dynamic weight adjustment and acquire a target push score matrix corresponding to the target user. If first target user data corresponding to an influencing factor satisfies a matching condition corresponding to the influencing factor, the influencing factor is taken as a correlation factor corresponding to the target user to acquire a correlation factor set corresponding to the target user. A sum of target push scores of all correlation factors in the correlation factor set corresponding to a candidate item is taken as a priority score corresponding to the candidate item. While retaining the global statistical advantage, the method significantly improves the personalization degree and the discrimination degree of the push, and enhances the discrimination degree and the accuracy of the priority of the candidate item corresponding to different users.

The invention discloses an abnormal state automatic identification and correction system for a radiopharmaceutical production line, and relates to the technical field of pharmaceutical production. The data collection module is used for collecting production line data and a data standard interval, and distinguishing an abnormal data set and a pseudo normal data set; the data processing module is used for judging the number of associated samples of the abnormal sample data; obtaining a net normal data set; establishing data type pairs, calculating correlation factors of the data type pairs, performing grade division and grouping on the data type pairs, and calculating a difference threshold value of each group of data; and calculating a predicted abnormal value. According to the method, by associating the sample number, obtaining the net normal data set, eliminating the influence of data fluctuation before abnormality, and comparing the difference index of the new data with the difference threshold value, when the abnormality index of the new data exceeds a certain range, it is judged that the production line data is likely to be abnormal, the prediction effect is achieved, and correction is conducted in time; and possible abnormal conditions are pre-judged, so that the occurrence of abnormal states is reduced.

The invention provides a power equipment control method and system based on the Internet of Things, and relates to the technical field of power equipment operation and maintenance. The method comprises the following steps: acquiring historical equipment state data, extracting multiple groups of temperature regulation reference data, extracting temperature regulation characteristic values and characteristic change values of multiple parameters, and calculating to obtain a temperature regulation correlation factor of each parameter; multiple target parameters are determined, a training data set is extracted from the multiple sets of temperature adjustment reference data, and a temperature adjustment analysis model is obtained through training; target equipment state data of the target transformer are collected, first target feature values of the multiple target parameters and change parameters of each first target feature value are extracted, multiple second target feature values are obtained through calculation, and a target score is generated through a temperature adjustment analysis model; and when the target score is greater than a preset early warning threshold value, performing temperature adjustment on the target transformer based on a preset temperature adjustment mode. According to the invention, the operation stability of the transformer in a complex operation environment is improved.

The invention relates to a user portrait analysis method and system based on a big data model and a medium, and the method comprises the following steps: collecting voltage data, and integrating user basic information and power grid topological structure data; constructing a user voltage prediction model; classifying according to the voltage data of the users in four seasons in one year; establishing an association relationship among temperature change, workday and weekend multi-dimensional association factors and user voltage, and obtaining an influence weight coefficient of each factor; carrying out k-means clustering analysis; generating a detailed voltage portrait for each user or region according to the analysis result; according to the voltage portrait, a power grid enterprise takes measures in a targeted manner to improve the voltage quality of a user and the power supply reliability; through continuous data collection and analysis, the user voltage portrait is updated in real time. According to the method, the user voltage is decomposed into the basic voltage and the sensitive voltage according to seasons, clustering analysis is carried out on the basic voltage and the sensitive voltage, and the portrait analysis result obtained through clustering is more accurate.

The invention provides a candidate item priority determination method for a civil aviation system, and relates to the technical field of data processing, and the method comprises the steps: firstly obtaining a basic push score matrix according to a historical data set, and then modifying the basic push score matrix in combination with a score adjustment weight matrix corresponding to a target user, realizing dynamic weight adjustment to obtain a target push score matrix corresponding to the target user; if the first target user data corresponding to the influence factor meets the matching condition corresponding to the influence factor, taking the influence factor as an association factor corresponding to a target user to obtain an association factor set corresponding to the target user; the sum obtained by adding the target push scores of all the correlation factors in the correlation factor set to the candidate items is used as a priority score corresponding to the candidate items; while the global statistical advantage is reserved, the personalized degree and the distinction degree of pushing are remarkably improved, and the distinction degree and the accuracy of the priorities of the candidate items corresponding to different users are enhanced.

The invention discloses an IQ imbalance calibration method and system. The method comprises pre-calibration and fine calibration. During pre-calibration, firstly, M frequency points are selected in a sectional mode, narrow-band calibration is carried out, correlation factors of amplitude imbalance errors and phase imbalance errors of the multiple frequency points are obtained, interpolation fitting is carried out on zero frequency points, and frequency response characteristics in the whole bandwidth are obtained. On the basis, a solution equation is constructed to calculate a filter coefficient, and a filter is designed to compensate the IQ imbalance error. During fine calibration, based on the least square method, the compensation coefficient of the filter is calculated in a real-time and self-adaptive manner, whether iteration is updated or not is judged, iterative filtering is carried out to enhance the stability of the system, and the problem of calibration performance deterioration caused by environmental factors such as temperature change is effectively solved. Compared with traditional narrowband calibration, the calibration system combining pre-calibration and fine calibration has a calibration effect on receiver large-bandwidth IQ imbalance, the adaptive capacity of the calibration system to environment temperature changes is enhanced, and the calibration precision and robustness are improved.

The application relates to a daily power consumption prediction method based on a fuzzy membership function, which comprises the following steps: step S1: constructing a time sequence of historical daily power consumption and historical air temperature by taking date as an index; step S2: establishing a fuzzy membership function; step S3: taking the maximum Pearson correlation coefficient as an objective function, and optimizing the parameters of the fuzzy membership function through a meta-heuristic algorithm; step S4: converting the historical air temperature and the predicted air temperature into relevant factors based on the optimized fuzzy membership function, and marking the relevant factors as covariates; and step S5: establishing a time sequence prediction model, training the model by using the historical daily power consumption and the historical covariates, and finally inputting the predicted covariates to calculate the daily power consumption prediction value. The application fully utilizes weather forecast data, and effectively improves the accuracy of daily power consumption prediction of a power grid.

The invention discloses an energy consumption anomaly analysis method and system for the environmental sanitation industry, and belongs to the technical field of environmental sanitation energy consumption analys.The energy consumption anomaly analysis method for the environmental sanitation industry comprises the following specific steps that firstly, multivariate data is collected in real time, and energy consumption time sequence data of a target area are collected and comprise the water and electricity consumption and the vehicle oil consumption value; step 2, synchronizing correlation factor data, including people flow, in-use equipment number, weather influence factors and similar area reference data vectors; according to the method, factors such as meteorology, people flow and equipment are dynamically coupled through a multi-modal feature fusion function, environmental interference is eliminated, the anomaly detection accuracy is greatly improved, a large language model outputs an actionable conclusion, the anomaly date is accurately locked, the correlation factor contribution degree is analyzed based on an attention mechanism, and the anomaly detection accuracy is improved. And outputting three interpretable labels of equipment fault, data false report and environment abrupt change to guide targeted disposal.